Atrial fibrillation in the ICU: Tough AF to Treat, Sick AF to care for…

Peer Reviewed by Dr. Sagar Dave, DO

Every once in a while, it’s a good thing to get down to the nitty gritty of things. I think this is most helpful in common situations. Atrial fibrillation is one of those things to have quickly accessible in your internal brain. In this episode I want to go through some useful studies regarding atrial fibrillation (AF) in the ICU. As usual this is not a how to guide but my synthesis of select articles regarding atrial fibrillation in the ICU. As such this blog is not intended to substitute for medical knowledge by an experienced provider so this is not an authoritative review. It is intended to be an informational supplement that was briefly peer reviewed.  I divided this up into 3 sections. The first is the real down and dirty nitty-gritty painful detail. This is followed by a brief organized summary section ending with an even more abridged algorithm in an attempt to distill it all into a cohesive thought. So, read ahead with caution!  


  • AF Atrial fibrillation
  • CI: Confidence Interval
  • NOAF: New Onset AF
  • AA: Anti-arrhythmic
  • HR: Hazard ratio
  • RR: Risk ratio
  • DCV: Direct Cardioversion
  • EF: Ejection Fraction
  • OR: Odds Ratio
  • NE: Norepinephrine



Causes of AF 

AF risk factors in the critically ill have not been consistently linked to traditional risk factors associated with AF in the ED or community setting (ie, structural and valvular heart disease). It is thought that acute events during critical illness accelerate cardiac remodeling and fibrosis to rapidly produce a susceptible atria (the so-called “atrial substrate”). This allows for the development of sustained AF in as a result of the assault on the body in the ICU by a variety of arrhythmogenic triggers. Accelerated remodeling can occur due to infection and inflammation. Murine and primate models of pneumonia show that bacteria deposit within the myocardium and result in development of atrial fibrosis eve with the treatment with antibiotics. Furthermore, bacteria can alter ion channels gene expression through toxin release [Bosch et al., 2018]. In addition to the bacteria and response to critical illness, we in the intensive care unit further flagellate the heart. Dopamine and epinephrine in particular have chronotropic effects that can lead to increased atrial ectopic discharges triggering new AF. 

 Recall dopamine, in particular in the 2010 SOAP II Trial[a], almost doubled the rate of atrial fibrillation in septic patients (20% vs 11%) [De Backer et al., 2010]. 

Greater illness severity is also associated with the risk of new AF development. Lastly, atrial size on echocardiography is associated with new-onset AF in the ICU, suggesting that iatrogenic atrial pressure/ volume overload may also be important in the development of AF in the critically ill [Bosch et al., 2018].

Risk Factors Associated with AF

The above figure is from a study by Kanje and shows risk factors identified at the onset or immediately before the development of new-onset AF (n = 139)[Kanji et al., 2012].

In a study by Moss of 8356 ICU patients, 10% had new onset atrial fibrillation (NOAF). The strongest associations were acute respiratory failure, advanced age (> 60 yr), and sepsis[b] [Moss et al., 2017]. Weaker but still significant associations were postoperative state, severity of illness, hemorrhage, vasopressor requirement, valvular heart disease, gender, and chronic lung disease [Moss] [Ibid.]. Heart failure, kidney disease, and body mass index (BMI) were not significantly associated with NOAF [Ibid.].

Incidence of NOAF

Walkey[c] in 2011 published a retrospective cohort of a California Database of administrative claims -over 3 million hospitalized adults. Severe sepsis (n = 49,082) occurred in 1.56% of those hospitalizations. New-onset AF occurred in 5.9% of patients with severe sepsis vs 0.65% of patients without severe sepsis [Walkey et al., 2011].

In another study by Walkey in 2014 they identified138,722 sepsis survivors via a Medicare 5% sample. In that group 7% (9,540) had NOAF during sepsis, 24% (33,646) had prior AF, and 69% (95,536) had no AF during sepsis. AF occurred following sepsis hospitalization more commonly among patients with NOAF during sepsis (54.9%) than in patients with no AF during sepsis (15.5%) [Walkey et al., 2014].

In 2020, a study by Fernando found 10% of admissions had NOAF. 22.4% of those NOAF patients went on to have sustained AF lasting longer than 24h [Fernando et al., 2020]. 

Furthermore, we may be missing some AF. Moss performed a retrospective cohort of ICU admissions for NOAF using automated detection (≥ 90 s in 30 min). In 8356 ICU admissions there were 123 (1.5%) documented cases of NOAF. However, they found sub-clinical AF in 626 patients (7.5%) for an overall incidence of any NOAF of 9%. Furthermore, in the Moss study, only 123/749 (16%) were likely persistent [Moss et al., 2017]. 

I’m sure as time goes on and more elderly come into the ICU and technology advances to have minute by minute recordings of vital signs the incidence of AF will continue to rise. 

Hemodynamic instability and AF

This will be the hand waving portion of the day. It is VERY difficult to distinguish instability of AF vs instability of critical illness. In the ICU, incredibly sick patients are “stable” on 0.25 (or more!) mcg/kg/min of NE, 0.03 of epi, 0.03 of vaso and have gotten 4L of fluid.  We have to figure out if there is a driving force that is making the patient unable to come off support and is that driving force AF. In that sense, I submit that “stable”—critically stable or meta-stable—p atient in the ICU has a VERY different connotation (as opposed to denotation) than many might be used to. 

Interestingly, 37% of critically ill patients with NOAF developed immediate hemodynamic instability, 11% exhibited new signs of cardiac ischemia and heart failure [Bosch et al., 2018].

In a study by Kanji et al. those with unstable AF were more likely to be receiving vasopressors/inotropes at the time of AF onset (53% vs 29%), to have decompensated heart failure resulting in pulmonary edema within 24 hours before the onset of AF (18% vs 6%), or to have an initial ventricular response greater than 150 beats per minute within the first 6 hours of AF (35% vs 16%, P = .01) [Kanji] [Kanji et al., 2012]. 

For this reason, maybe looking for new heart failure, a faster rate and increasing need for vasopressor might be helpful (or might not…).

Rates of attempted cardioversion of AF during critical illness are low: in postoperative patients who developed AF, attempted cardioversion resulted in immediate conversion to sinus rhythm (SR) in 71% of patients, but after 1 h, only 43% patients remained in SR, and after 24 h, only 23% patients remained in SR [Bosch] [Bosch et al., 2018]

Side Bar: What is your MAP goal?

Typically, we (stringently) target a goal MAP of 60. However, can a lower MAP goal be tolerated in critically ill patients? While this is subject of much debate one author looked at the elderly patients and the use of a lower MAP goal (60-65mmHg)

Lamantagne performed a multicenter, open-label feasibility RCT of 188 patients and compared lower (60–65 mmHg) to higher (75–80 mmHg) MAP targets and found no difference in SECONDARY outcomes of mortality (28d) or Renal SOFA scores [Lamontagne et al., 2016].

This prompted a follow up RCT by the same investigators.  “The 65 Trial” randomized 2600 patients over age 65 to MAP >60 mmHg or >65mmHg. They found no difference in death (41% vs 44%) and no difference in CRRT (24% in both groups) or urine output. However this endpoint did not lower supraventricular arrhythmias (12 vs 13 episodes) [Lamontagne et al., 2020]. 

[See Blog post:

Side Bar: Is your [radial] arterial line accurate? 

I love putting in a good arterial line but often you will see numbers you cant interpret[d]. Thus, we end up comparing it to something else (like a BP cuff and BTW the MAPs should be within 10 of each other). From the two studies below maybe we should switch out that radial for a femoral arterial line in shocked patients with up trending vasopressor requirements? (A maximal intensity suggestion, I know)

One prospective observational study of 159 patients in septic shock compared simultaneous arterial measurements of radial and femoral lines. Mean difference between radial and femoral MAP was +4.9 mmHg; during high-dose NE (>0.1 mcg/kg/min) NE this increased to +6.2 mmHg (95% CI: -6.0 to +18.3 mmHg). MAP differences > 5 mmHg) occurred in up to 62.2% of patients with high-dose NE therapy [Kim et al., 2013].

A second prospective observation trial of 77 patients in septic shock also comparing simultaneous radial and femoral lines found a difference in 75.4% of cases up to 5 mmHg. Of those, 25% had gradients more than 5 mmHg, 20.4% had the femoral MAP greater than radial MAP and 4% had radial MAP > Femoral MAP. The interval (difference in no direction) ranged from 16 mmHg in the no-NE group, 18 mmHg in the low-NE group, and 19 mmHg in the high-norepinephrine group [Antal et al., 2019].

Is AF a harbinger of Acute MI or Pulmonary Edema

In the study by Kanji [Kanji et al., 2012] of 3081 patients over 1 year in the ICU they found 348 patients (10%) with AF. 4.5% of them had NOAF and 6% had preexisting AF. Acute myocardial infarction occurred for the first time in the 24 hours after AF in 7% (9/139) of the NOAF group and in the preexisting AF group in 4% (8/186). Acute pulmonary edema occurred for the first time in the 24 hours after AF in 4% (6/139) of the NOAF group and in the preexisting AF group in 2% (4/186).

Thus, while a small incidence it is something to consider and not just focus on the rhythm! I’m not sure we need to get troponin on every patient but I do think we take a good look at those ST segments on the ECG and even another reason to not settle for a monitor produced ECG.

Treatment Options

Obviously, this is not an exhaustive list. This is going to be a review of MY go to strategies. As always in medicine there are many options. This is just one person’s journey…

Should we treat atrial fibrillation at all (Do you even cardiovert, brah?!)

Before we actually talk about treatment, do we even need to treat atrial fibrillation in the sick patient? Obviously no one knows the answer to this. From above it does seem some of this is transient and never even diagnosed. However, in a retrospective descriptive cohort study from 2 urban ED’s Scheuermeyer looked at ED patients with atrial fibrillation and an “acute underlying illness” [Scheuermeyer et al., 2015]. They identified patients 416 patients who were divided into those treated for the underlying condition with AF and those treated only for the underlying condition. 135 had rate and/or rhythm control; 281 were not treated for AF  (35% of the treated and 30% of the not treated were diagnosed with sepsis). Of the 135 with AF control 19% had successful rate control and 13% had successful rhythm control. 40% of the treated for AF group had any adverse events; vs 7.1% of those not treated for their AF. Adverse events were divided into major and minor events:

  • 14% of the treated vs 1% of the untreated had major adverse events: hypotension requiring pressors and Intubation
  • 33% of the treated vs 7% of the untreated had minor events: hypotension requiring fluid bolus and bag-valve-mask oxygenation

Direct Cardioversion (DCV)

I love me some electricity! I’m comfortable doing it and we know in the ED setting it is safe and likely associated with shorter ED stays [Stiell et al., 2020]. Since ICU NOAF is not as well studied in the literature it is difficult to distinguish success of amiodarone vs DCV because they are so often used together. So this water is likely pretty muddy and DCV and amiodarone are entangled.

In a study by Kanji, DCV was attempted in 26 (19%) of 139 patients with new-onset AF, (70% had received amiodarone just prior or during DCV). Conversion to sinus for ANY duration occurred in 13/26 (50%) but was maintained for at least 24 hours in only  (27%) of 7/26. By comparison Conversion to sinus for ANY duration in the amiodarone group occurred in 103/116 (88%) but maintained for at least 24 Hours in only 24/116 (20%) [Kanji et al., 2012]. A low yield of DCV appears to be consistent in the only other study of DCV in critically ill. 

Mayr performed DCV in 37 patients. 13 patients (35%) converted to SR, of those 8 patients remained in SR (24%) at 1 hr, 6 patients (16%) at 24 and 5 patients (13.5%) at 48 hrs. Notably, this study used monophonic waveforms and anterolateral placement [Mayr et al., 2003]. It would seem if you are going to cardiovert, do it as early as possible near the onset as success seems to decrease. 

In the category of most unexpected, Blecher found that in cardioversion of AF in the ED, drug use PRIOR to electrical cardioversion reduced the success of electricity. This is OPPOSITE of what is commonly believed that slowing the ventricular response before attempting cardioversion increases the success rate, although there is little evidence to support this. Blecher (part of the Ian Stiell research powerhouse) looked at ED cardioversion for discharge home in 634 patients who underwent attempted cardioversion: 428 electrical, 354 chemical, and 148 required both. They had 378 successful and 50 unsuccessful electrical cardioversions. Medications used included: Beta-blockers (~33%), Sotalol (~10%), Amiodarone (~5%) and Digoxin (~2%). They found that 64% of those failing electrical cardioversion had the ventricular rate slowed before the attempt at electrical cardioversion versus 38.4% of those successfully converted. Thus, more failures at electrical cardioversion had rate control prior! They found rate control and prior attempted chemical cardioversion was associated with decreased likelihood of successful electrical conversion: OR 0.39 (95% CI 0.21–0.74) and 0.28 (95% CI 0.15–0.53)   [Blecher et al., 2012]. A clinically significant OR (Odds Ratio) is considered to be an OR 0.3 or less!

Electrode Positioning in DCV:

In (what I could find as) the only RCT of electrode positioning a study by Kirchhof[e] et al. of PERSISTENT AF, cardioversion was successful in a higher proportion of the anterior-posterior than the anterior-lateral group ([96%] vs [78%]). Cross-over from the anterior-lateral to the anterior-posterior electrode position was successful in 8/12 patients, whereas cross-over in the other direction was not successful (0/2) [Kirchhof et al., 2002].

Other observational trials have shown inconsistent results regarding electrode positions on the success.  In a meta-analysis of 10 trials with 1281 patients the anterior-posterior electrode position had no advantages in terms of success of electrical cardioversion [Zhang] [Zhang et al., 2014].

However, even in this trial they noted the only study without bias is the Kirchoff [Kirchhof et al., 2002] trial so I continue to use the AP position. Also, the AP position to me makes sense in CARDIOVERSION since you want to be across the atria, the anterior lateral position makes sense to me in DEFIBRILLATION since you are across the LV. Impedance of the chest wall is likely a large factor and is likely affected by BMI. You can ask me in person my strategy to overcome high chest wall impedance. It’s shocking!

Magnesium (Mg)

Don’t you wish sometimes we still lived in that blissful world of medical school and board exams where all cases of appendicitis are in the right lower quadrant, zebras roam freely, and all hypokalemia and hypomagnesima are the cause of AF?! Sadly not even the latter statement is true. 

Lancaster Incidence of AF does not correlate with Potassium (K) or Magnesium (Mg) serum levels. See Text

Lancaster in a study of 2041 post-operative AF (POAF) patients without pre-operative AF looked at just this idea. They found that in 752 patients with POAF patients had higher potassium and magnesium levels than matched patients POAF. Further more they found K, Mg supplementation did not reduce the rate of POAF [Lancaster et al., 2016].

It should be clear that cardiac surgery patients are a slightly different brand of AF but I wouldn’t be surprised to find this is true in the ICU as well. Nevertheless, I will continue to write K>4 and Mg>2 in my notes. 

Although it may be a surprise I do still believe that magnesium is capable of cardioversion in AF. A meta-analysis of magnesium for the prevention of postoperative atrial fibrillation in cardiothoracic patients found an odds ratio of 0.66 (95% CI: 0.51 – 0.87) [Henyan et al., 2005].

In a prospective single arm trial in critically ill patients a “step up” protocol by Sleeswijk of magnesium followed by amiodarone was shown to improve the rate of cardioversion [Sleeswijk et al., 2008]. Sadly this trial looked at only 26 patients. A Mg bolus (0.037 g/kg over 15 minutes) followed by infusion (0.025 g/kg/h) was given to all patients with persistent AF DESPITE correction of K, or Mg. While this is the equivalent of 4g of magnesium in a 100 kg patient, it should be nothing to be concerned about as this is a safe dose in patients with no renal disease and given frequently to OB patients[f]. The INFUSION was cut in half when Mg was >2.0 mmol/L and stopped when Mg >3.0 mmol/L. If no conversion in 1 hour then amiodarone (300 mg over 15 minutes, and infusion of 1200 mg/24 h) was started. The results:

  • 16/29 (55%) patients responded to magnesium alone
  • 11 of the remaining 13 responded to the addition of amiodarone 
  • 27/29 (93%) responded to the combination [Ibid.].

This study looked at the equivalent of 4 g OVER 15 min. Certainly nothing to freak out over. Previous studies have shown a benefit of Mg in patients with AF by increasing the success of pharmacological cardioversion and by decreasing the incidence of postoperative AF [Rajagopalan et al., 2016]. Sadly, the data for magnesium is not of the highest quality. 

Rajagopalan performed an RCT of chronic AF patients showed no difference in cardioversion with or without magnesium (86%). These patients were very different and had chronic AF and were in AF for 3-4 months prior. Even after 4 shocks up to 200J only 86% of these patients cardioverted to SR! As a side note in my favor 2/132 in the mag group converted to SR prior to cardioversion vs none in the placebo group. They found no drop in BP with the magnesium and no adverse events [Ibid.]. This information with other the studies makes me hopeful. 

In my opinion, I like using magnesium especially if it gives me 30 minutes to decided what I want to do and do a second look at the patient! 

Amiodarone for AF

Amiodarone is a very old drug and thus there are numerous different dosing recommendations. 

PDR: Adults Intravenous dosage recommendations [PDR].

  • Initial IV rapid infusion of 150 mg over the first 10 minutes. 
  • And Then 1 mg/min for the next 6 hours (total dose infused = 360 mg).
  • And Then, the infusion rate is lowered to 0.5 mg/min for the next 18 hours (total dose infused = 540 mg). 
  • And Then after the first 24 hours, a maintenance IV infusion of 0.5 mg/minute (720 mg/day) is recommended.
  • No and then! Intravenous amiodarone should not be administered for longer than 3 weeks.  
  • NOTE: The dose of amiodarone may be individualized

Conversion from intravenous to oral therapy [Goldschlager et al., 2000]:

  • If duration of IV infusion was <1 week, the initial oral dose is 800 to 1200 mg/day PO.
  • If 1 to 3 weeks, the initial oral dose is 400 to 800 mg/day PO. 
  • If longer than 3 weeks, the initial oral dose is 300-400 mg/day PO.
  • If there is concern about GI function, both oral and IV therapy should be maintained for a few days

Amiodarone Pharmacology and kinetics (Skip this unless you are feeling particularly nerdy) 

Amiodarone contains 37.3% iodine by weight. It is a Vaughan Williams (remember those) class III antiarrhythmic (AA) but produces activities with each of the 3 other classes as well. Other pharmacologic activities include: systemic/coronary vasodilation, phospholipase inhibition, and inhibition of thyroid hormone metabolism. N-desethyl- amiodarone (DEA), the major metabolite also has antiarrhythmic activity. The half-life of amiodarone is 20-47 days and that of DEA is even longer [Chow, 1996].

Amiodarone Side Effects

In adults, IV (not oral) amiodarone dosage adjustments are not required on the basis of patient age, or renal or hepatic functionThe likelihood of pulmonary fibrosis with short-term use of intravenous amiodarone appears small.There is a reported 3.4% incidence of hepatic enzyme elevations combined with at least possible hepatic dysfunction with IV amiodarone thus it is important to monitor hepatic function. Most studies exclude patients with thyroid dysfunction and it is recommended to check thyroid function once and then at 6 months if the patient is still on amiodarone [Ibid.].

Effects on Thyroid function 

Amiodarone induced hypo/hyper thyroid (AIH) and amiodarone induced thyrotoxicosis can occur (AIT). The prevalence of AIH is as high as 22%. Acutely, there is an increase in TSH (but usually <20 mU/L), an increase in both free and total T4, and a decrease in total and free T3. After 3 months, a new equilibrium is reached, and TSH normalizes. However, T4 will remain high, and T3 will remain low. It is best to avoid checking thyroid function tests during the first 3 months of treatment. Most patients who do not have underlying Hashimoto’s thyroiditis will have resolution after amiodarone is stopped. The prevalence of amiodarone-induced thyrotoxicosis (AIT) is much lower than AIH. AIT can occur quite suddenly and at any time during treatment. AIT is diagnosed based on a suppressed TSH with an elevated free T4. Given the beta- blocking effects of amiodarone, the classic findings of thyroxicosis are often absent. In equivocal cases, a T3 level can be helpful, with an elevated or high normal T3 indicating thyrotoxicosis [Goldschlager et al., 2007].

Pearl: Clinically, the most common findings may be weight loss or a change in warfarin dose.

Amiodarone Mechanism of Action

Interestingly, short term mechanism of action (single dose) mainly produces AV nodal refractoriness and prolongs intranodal conduction interval time, (Class II and IV).  The long-term activity is an increase in the action potential duration in cardiac tissues (Class III). Negative inotropy is the most consistent hemodynamic effect of IV amiodarone. In patients with normal ejection fraction (EF), negative inotropy is usually offset by a decrease in systemic vascular resistance to maintain cardiac output. Thus, patients with left ventricular dysfunction are at greater risk for decrease in cardiac output but this is likely only transient [Kosinski et al., 1984].

Amiodarone Dosing

I find that dosing of amiodarone to be incredibly provider dependent and that is likely due to the extreme variations in patient response to this drug. The pharmacokinetic reason for this is that lasma concentrations of the drug do not correlate well with observed clinical effect because of rapid distribution to tissues and high plasma protein binding [Desai et al., 1997].

A study by Kosinski showed this on the effects of a 300-mg bolus dose over 5 min, followed by a continuous infusion (1000 mg/24 hover 3-5 d) in 12 patients. Patients with the higher EF (>35%) had a small but significant increase in cardiac index following whereas patients with the lower EF (<35%) had a decrease in cardiac index (from 2.1 to 1.7 L/min), MAP and increase PA pressures which after 3-5 days were compensated for by peripheral vasodilation (decreased SVR). Two (out of 6) patients in the low EF group developed hypotension requiring pressors. They recommended a longer infusion for the 300 mg dose in low EF patients [Kosinski et al., 1984]. 

In 1983 a study by in Mexico, Faniel looked at 26 patients admitted to the ICU SPECIFICALLY FOR AF with RVR. They used 3 mg/kg (1983 weights as well) over 3 min vs 5-7.5 mg/kg over 30 min for a total of 1500 mg in 24 hours. 5 were unsuccessful and required cardioversion (4 of these 5 had HR slow to 50 bpm as reason for failure). Mean HR was 140 (55-200). The initial dose was followed immediately by a slow infusion of 600 to 1200 mg/24 h to reach a maximum of 1500 mg administered during the first 24 h. Mean conversion time was 170 min. They saw no hypotension. They noted “The longest reversion times were generally related to situations where small, repeated initial doses had been given. This mode of administration appears less effective than giving the same total amount in a single dose” (175 min vs 240 min, larger vs smaller boluses). They concluded: even if stable reversion is not achieved, then DC shock may be improved by the prior use of amiodarone (we just read above this may not be true). Overall they concluded that there was a good hemodynamic tolerance to their dose of 7 mg/kg over 30 min [Faniel and Schoenfeld, 1983].

In 2004, Hofmann looked at 78 AF patients in a CCU with advanced congestive heart failure or cardiogenic shock (SBP<90). 13 required pressers none were on mechanical circulatory support. Patients were given a single bolus of 450 mg of amiodarone via a PIV (over 1 min?). cardioversion was successful in 40 patients (51.3%) within 24 hours: sinus rhythm occurred in 25 patients (32%) within 30 minutes after amiodarone, and during the following 23.5 hours another 15 patients (19%) reverted to sinus rhythm. They noted “ In two patients, a decrease of systolic blood pressure from 115 to 80 mmHg and from 130 to 100 mmHg occurred within the first 5 minutes, but blood pressure returned to the initial values after 10 and 90 minutes respectively without specific intervention”[Hofmann et al., 2004].

Then in 2006, Hofmann looked at 50 consecutive AF with RVR patients and compared to amiodarone to digoxin. In this trial they excluded patients with SBP <100 and preserved Ejection Fraction (EF). They also looked at 450 mg of amiodarone via a peripheral IV (PIV)[g].  This time specifically stated as a bolus over 1 min. If the ventricular rate was above 100 bpm after 30 min, patients received another 300 mg IV. 28 patients required a second dose of amiodarone. Sinus rhythm conversion occurred quicker in the amiodarone group. SBP fell about 10 mmHg on average after amiodarone. 4 patients require fluid bolus. No prolongation of the QTc interval occurred [Hofmann et al., 2006].

AF: Amiodarone vs All comers

There are not a ton of great studies of drug choices and septic shock

A study by Balik looked 234 patients with septic shock requiring NE for propafenone vs amiodarone vs metoprolol. There were 177 in the amiodarone group, 42 in the propafenone group and 15 in the metoprolol group. The cardioversion rate was: 74% with amiodarone, 89% with propafenone, and 92% with metoprolol. The 28 day mortality was 50% in the Amiodarone group, 40% in the propafenone group, 21% in the metoprolol group. Multivariate analysis demonstrated higher 12-month mortality in amiodarone than in propafenone.  (HR 1.58 95% CI:1.04-2.38; p = 0.03) [Balik et al., 2017].

In this study by Delle Karth of amiodarone vs diltiazem (I can’t leave my ER roots, Kate!) in critically ill patients, they looked at 3 groups of patients in the ICU of 20 patients with Apache scores of 75. Approximately 75% of the groups required mechanical ventilation and catecholamine therapy, thus these were sick patients. The groups were diltiazem (25 mg bolus + 20mg over 24 hrs), amiodarone bolus(300mg), amiodarone bolus + infusion (300 mg+ 45 mg/hr for 24 hours). The primary outcome was a 30% reduction in HR and a secondary outcome was a HR <120. Diltiazem allowed for a 30% HR reduction in 70% of patients and a HR <120 in 100% of patients. Amiodarone bolus only allowed for a 30% HR reduction in 55% of patients and and a HR <120 in 50% of patients. Amiodarone +infusion allowed for a 30% HR reduction in 75% of patients and and a HR <120 in 95% of patients. Hypotension occurred in 6/20 in diltiazem group and 0/20 in either amiodarone group [Delle Karth et al., 2001]. Thus it appears my ED favorite may cause more hypotension in these very sick patients than amiodarone. 

Chapman studied IV procainamide and compared it to Amiodarone in 24 critically ill patients. IV amiodarone (3 mg/kg followed by 10 mg/kg/24 h, with repeat dose of 3 mg/kg at 1 h if no response) or IV procainamide (10 mg/kg at 1 mg/kg/min followed by an infusion of 2-4 mg/min for 24 h, and a repeat dose of 5 mg/kg at 1 h if no response). The patients were recruited from a mixed medical and surgical ICU. Most were on mechanical ventilation (20/24), had sepsis (18/24) and an avg APACHE score of 21. They found conversion to sinus rhythm by 12 h in 10/14 (71%) in the procainamide group and (7/10) (70%) in the amiodarone group. SBP was not significantly different from baseline for either drug [Chapman et al., 1993]. I happen to be a big fan of procainamide in the ED for conversion of AF for discharge home, however, I have not had experience using it in the ICU due to lack of availability. 

Esmolol atrial fibrillation, and Mortality?

In a study of esmolol in patients with septic shock in TACHYCARDIA (avg HR 109) (not due to AF), Brown looked at 7 (yes, 7 patients!…because the other 179 patients were excluded)… Inotropes (not vasopressors) were one of many reasons to stop treatment [Brown et al., 2018]. A review article by Arrigo states “We recommend to start with substances with a low risk profile and short half-life, such as beta blockers (see below), and to escalate to other substance classes such as amiodarone only in cases of contraindications or inefficacy…Our choice is esmolol”. They cite the reason as “[beta blockers] significantly reduces the risk of AF up to 40%, particularly in the [cardiac surgery] postoperative phase”. The recommend a dose of 10–20 mg to reach 1 mg/kg Bolus. If the MAP is >60 mmHg, start infusion at a rate of 0.05 mcg/kg/min and may increase q 30-minute intervals as needed for HR. They recommended use, especially, if a patient was on oral BB prior [Arrigo et al., 2014]. Unfortunately, this review gave no references to support the use of esmolol for NOAF in ICU patients.

Another review article by Bosch also recommends Esmolol as FIRST LINE in AF in critical illness however they also have no evidence for this. They state “Thus, use of BBs to treat arrhythmias during critical illness is a promising area of investigation.” [Bosch et al., 2018].

Although these review articles did not give any evidence for the use of esmolol in AF and the sepsis syndromes, I think much of the reason for these recommendations for esmolol in critically ill patients with AF likely stems from the next two studies we will discuss.  

 An open label Italian  RCT by Morelli looked at septic shock patients WITHOUT AF! 77 patients were randomized to esmolol in septic shock and 77 to usual care. This was a feasibility study so the primary outcome was heart rate. Remember that secondary outcomes are hypothesis generating! They did indeed manage to keep the HR down for their primary outcome. Additionally, they reported a 28d-mortality of 49.4% in the esmolol group vs 80.5% in the control group [Morelli et al., 2013].  Naturally, in 2013 this study made headlines. However, lets check those numbers! That was an 80% mortality in the control group!!! Thus 30% difference… Interestingly they also reported about a 500 ml reduction in fluid administration to the esmolol group. Normally, I hear “Esmolol? That’s a lot of fluid [administered]!” Very interesting indeed. 

While there is no RCT data on AF, septic shock and beta blockers, there is a very large prospective observation trial looking at different rate controlling drugs and their mortality. This one even has a subgroup of septic shock! 

Once again, Walkey performed this interesting observational study of AF and sepsis in 2016. The study was a retrospective cohort of billing data from about 20% of US hospitals. Importantly, we have no idea why the clinician chose the rate control method they did. They looked at 39,693 patients with AF during the first 14 days of hospitalization for sepsis treated with only one rate control drug. They found 36% treated with a calcium channel blocker, 28 % were treated with a beta blocker, 20% with Digoxin and 16% with amiodarone. In a propensity-matched analyses, BBs were associated with lower hospital mortality when compared with CCBs (relative risk [RR], 0.92; 95% CI, 0.86-0.97), digoxin (RR, 0.79; 95% CI, 0.75-0.85), and amiodarone (RR, 0.64; 95% CI, 0.61-0.69). This was similar among subgroups of:  new-onset AF,  preexisting AF, heart failure, vasopressor-dependent shock, or hypertension. Patients with vasopressor infusion/shock were compared. 

  • BBs vs CCBs: shock: RR, 0.86; 95% CI, 0.79-0.94; no shock: RR, 0.98; 95% CI, 0.91-1.06
  • BBs vs Digoxin: shock: RR, 0.79; 95% CI, 0.73-0.86; no shock: RR, 0.80; 95% CI, 0.73-0.88
  • BB vs Amiodarone in shock: RR, 0.64; 95% CI, 0.59-0.69; no shock: RR, 0.73; 95% CI, 0.65-0.81

The percent of patients on vasopressors per drug group was: BB 29%, CCB  26%, Dig 44%, amiodarone 64% [Walkey et al., 2016b]. It should be noted that regardless of how well a propensity matched score is done it can’t isolate bias like an RCT and association does not mean causation, so this needs to be interpreted with caution. On the other hand, if we use the numbers from Walkey and assume a 27% vs 42% mortality in beta blockers vs amiodarone; that gives a Number Needed to Treat of 6 patients to prevent 1 death with amiodarone. Finally, either there is a signal of a mortality benefit with beta blockers or amiodarone is a marker for sicker patients. As an emergency medicine trained person my first thought is for CCB or BB but in the shocked patient many are too hypotensive and I end up having to use amiodarone.  


Digoxin slows heart rate by increasing vagal tone; it is associated with low rates of hypotension but has a narrow therapeutic index. Observational studies show associations between digoxin use and increased mortality. Vagomimetic effects of digoxin may be less effective during critical illnesses characterized by high catecholamine states [Bosch et al., 2018]. Digoxin should not be considered as a first-line option for rate control due to its slow onset of action [Sibley and Muscedere, 2015]. So…no matter what ANYONE[h] says, Digoxin for AF is still a very 1950s Treatment… Hence we wont discuss pharmacology or dosing. 

Does Anything make NOAF better?

McIntyre performed a systematic review and meta-analysis of 23 RCTs with excellent methods and low risk of bias. They found that patients who had vasopressin + catecholamine vasopressor had a lower incidence of AF than did patients not on vasopressin:

  • 24% (136/559) had AF in the catecholamine + vasopressin group
  • 33% (182/554) had AFin the catecholamine only group

They found a risk ratio of 0.77 [95% CI, 0.67 to 0.88] (not something thought to be clinically significant but still statistically significant. Sadly, this study showed no benefit for mortality with vasopressin [215/529 (41%) vs 222/520 (43%)] [McIntyre et al., 2018]. A whopping 40% mortality in this group!

Treatment Duration

            While no one knows how long to keep anti-arrhythmic medications going, Kanji showed 18% of NOAF patients and 62% of patients with preexisting AF who survived to ICU discharge left the ICU in AF [Kanji et al., 2012]. Bosch reported, 70% of patients with new-onset AF and 14% of patients with preexisting AF converted for at least 24 hours within the first 48 hours from the onset of AF[Bosch] [Bosch et al., 2018].

Anticoagulation vs. Stroke Risk

It is unknown if administering anticoagulation in critically ill patients prior to DCCV decreases the risk of thromboembolic events or if there is an optimal timing of anticoagulation prior to DCCV [Ibid.].

In the large study by Walkey in 2014, 138,722 sepsis survivors were identified in a Medicare database. In those with new-onset AF during sepsis compared with those with no AF during sepsis, the NOAF group had a greater 5-year risk of hospitalization for ischemic stroke (5.3% vs 4.7%; HR, 1.22) [Walkey et al., 2014]. That is a 0.6% absolute increase in stroke.

In 2011, Walkey looked at AF, severe sepsis and stroke in the US. In over 49k patients with severe sepsis and NOAF, they found that 2.6% (75/2896) had in hospital stroke as compared to the 0.6%  (306/46,186) without severe AF for an adjusted OR of 2.70 (95% CI, 2.05-3.57; P < .001) [Walkey et al., 2011].

In the Kanji study in Canada, 348 patients and over 2322 cumulative patient days of AF in the ICU, no patients had a documented embolic cerebrovascular event, whereas 5 (9%) of 58 patients who received systemic anticoagulation had a bleeding event that required interruption of anticoagulation and at least 1 blood transfusion [Kanji et al., 2012].

In a 2016 cohort study of 38,582 hospitalized patients with atrial fibrillation and sepsis, Walkey found bleeding events were increased among patients who received anticoagulation (1163 of 13 505 [8.6%]) compared with patients who did not receive anticoagulation (979 of 13 505 [7.2%]). However, those patients did not have a significantly reduced risk of in hospital stroke (1.4%) compared to those receiving of anticoagulation during sepsis compared to those who did (1.3%)[Walkey et al., 2016a]. So bleeding ~1.4% vs stroke 0.1%.

Some recommend starting anticoagulation if the AF is persistent for more than 48 hours [Sibley and Muscedere, 2015]. Others recommend in patients without contraindications to anticoagulation whose AF persists following hospital discharge, anticoagulation should be initiated if moderate to high risk [Bosch et al., 2018]. Currently there is no quality evidence or guideline to guide decision making for atrial fibrillation in the critically ill. Based on this data and the bleeding risk in ICU patients it would seem not surprising that some patients are not started on AC. In fact, in a study in Canada systemic therapeutic anticoagulation was prescribed for 16% (22/139) of patients with new onset AF and 19% (36/186) of patients with preexisting AF while in the ICU [Kanji et al., 2012]. 

Mortality and morbidity associated New onset atrial fibrillation NOAF

In a retrospective analysis by Fernando in 2020 of 6 years of a registry from two Canadian ICUs  10% (1541 of 15014) patients were found have NOAF. These patients did not have a STATISTICALLY significant higher mortality (37.4% vs 29.9%) than patients without AF (OR1.002, p=0.31).  However, NOAF was associated with higher hospital mortality among ICU patients with suspected infection (aOR 1.21 [95% CI 1.08–1.37]), sepsis (aOR 1.24 [95% CI 1.10–1.39]), and septic shock (aOR 1.28 [95% CI 1.14–1.44]). They did have a statistically significant longer ICU stay by 1 day range: [4-14] vs [2-9].

In the 2014 study by Walkey, a multivariable-adjusted hazard ratio compared with patients with no AF during sepsis, those with new-onset AF during sepsis had greater 5-year risks of death (74.8% vs 72.1%; HR, 1.04; 95% CI,1.01-1.07) [Walkey et al., 2014].

In the 2011 study by Walkey comparing patients with severe sepsis with NOAF and severe sepsis without NOAF; they found a greater risks of in-hospital mortality (56% vs. 39%) [Walkey et al., 2011].

In a study IN CHINA, Liu looked at mortality of whether or not patients with AF converted to sinus. They found 503 eligible patients, including 263 patients with no AF and 240 patients with NOAF.  Of the 240 patients with AF, SR was restored in 165 patients, and SR could not be restored in 75 patients. The NOAF that stayed in AF group had the highest in-hospital mortality rate of 61.3% compared with the NOAF that converted to sinus group (26.1%) or the no NOAF (17.5%) group.  Interestingly the group that stayed in AF had higher baseline SOFA scores, APACHE scores, more norepinephrine use, more mechanical ventilation use and more dialysis. The group that has NOAF and stays in AF is likely a sicker group! Also note that a 61% mortality is really high (hopefully this was not confounded by an invisible virus!) [Liu et al., 2016]

Special Populations: Cardiac ICU and Lung Transplant

Until now I have mostly avoided discussing post-operative AF in cardiac surgery patients. I would go through the specifics of post cardiac surgery patients but the following review article does a way better job:

Lung Transplant Patients

Twenty-five per cent (25%) of lung transplantation patients developed atrial flutter or fibrillation, most frequently at day 5–7 post lung transplantation, and more commonly present in older recipients and those with underlying chronic obstructive pulmonary disease (COPD), but not in those with previously noted structural heart disease, or in those undergoing single rather than double lung transplants. Diltiazem can increase tacrolimus concentration and the lung transplant team should be notified to adjust the tacrolimus dose whenever diltiazem is started or stopped [Barnes et al., 2019].

Putting it all together….

So what to do, what to do…

So how do we put all this together? Keep in mind this is MY version of putting it all together (see the algorithm), there are lots of permutations and possibilities. I will also throw back a little Toxicology Bombastus[i] style that THERE IS NO BAD DRUG and only the dose makes the poison!  Well, let’s start at the beginning footnote always good place. Here, again to be sure, we are talking about the ICU patient who is very sick and develops new onset atrial fibrillation. Let’s use a sample case presentation:


Mr. Sic AF presents to your ICU and is admitted via the ED in septic shock. After 24 hours of treatment he is on 0.1 mcg/kg/min of norepinephrine, his map is 65 via a radial arterial line,  his HR is 115 and there is sinus rhythm on the monitor. He suddenly goes in to AF with a rate of 140 and his MAP is 61 now…what do you do wildcat (see corresponding algorithm)?!


  1. IF a patient becomes grossly unstable obviously resuscitate them.
  2. Get a real ECG if they are “stable” (see definition of stable)
    • Be a cardiologist here. Don’t substitute a monitor produced one, not one without a rhythm strip…An honest to god pink shiny papered ECG
  3. Is this rhythm causing additional instability more so than if they did not have AF? 
    • Is this tachycardia (AF or sinus) compensatory for the shock state as opposed to pathological and contributing to the current hemodynamic compromise
    • The atria supply about 25% of cardiac output and myocardial depression is common in sepsis especially in those who die of septic shock [Jardin et al., 1999]
    • Remember that unstable AF is more likely (but not impossible) to present with a HR of >150 (35% vs 16%)
  4. Do they NEED cardioversion?
    • Only cardiovert if you really think they are unstable
    • If you are going to cardiovert your best chance of success is in the first hour of AF. 
    • If you are going to cardiovert use the Anterior posterior placement
    • Remember only about 50% of cardioversion is initially successful and only about 1/3 successfully stay in sinus for 24 hours.
  5. Does this episode of AF need treatment?
    • Risk factors include age>60, chronic lung disease and sepsis [Fernando et al., 2020], [Moss et al., 2017].
    • Treat the underlying illness first if you think the high HR is due to critical illness then consider if the NOAF needs to be treated 
  6. Have the modifiable risk factors been sufficiently optimized?
    • Did you look at the ventricular function with bedside Echo?
    • Is there volume overload that can be optimized?
    • What are the electrolytes?
    • Are they acidotic?
    • Pressors: Avoid epinephrine and, especially, dopamine if possible
    • Can the pressor dose be titrated down without loss of blood pressure goals?
    • Can a lower MAP goal of 60 mmHg be tolerated [Lamontagne et al., 2020].
    • Is that [radial] a-line accurate? [Kim et al., 2013], [Antal et al., 2019].
  7. Give Magnesium[j]
    • I like to have a conversation with my local pharmacists, they are a wealth of knowledge and will let you know what your local hospital “policies” are as far as administration.
    • A small number of patients (but as high as 50%) might convert with this alone.
    • Give about 2g over 15-30 min (not 2 hours!) (consult your local hospital “ethereal” policy).  
    • If you want to give more magnesium then check repeat levels magnesium in a few hours (2-4 hours?)
  8. Choose and administer your anti-arrhythmic of choice
    • There are lots of choices here for the patient and treatment is usually based on personal preference. 
    • In a “stable” blood pressure diltiazem and esmolol are reasonable choices.
    • The aforementioned studies have looked at beta blockers while norepinephrine was infusing but not inotropes) 
    • observational data indicates that beta blockers might have a mortality benefit over amiodarone and digoxin (less so vs. calcium channel blockers) [Walkey et al., 2016b]. 
    • This needs yet to be confirmed in an RCT.
    • In the shocked patient who is on an inotrope or can not risk having more hypotension, amiodarone may be necessary.
    • Consider bolus of 150-450mg over 30 min then an infusion. See the algorithm for dose details.
    • The literature would suggest a 70% of patients but this will decrease by 24 hours and it will be mostly rate controlling. 
    • Amiodarone can initially WORSEN LV function and cardiac output but this will be offset in a few days by a decrease in SVR.
  9. Does the patient need to be anti-coagulated?
    • It would seem from above that less than 20% of patients are given AC to NOAF
    • These patients have a high bleeding risk (9% of those given AC have significant bleeding) so calculate a HAS-BLED score
    • It’s unclear that AC (vs prophylaxis dose) anticoagulation is helpful [Walkey et al., 2016a].
    • Think about it carefully then sleep on it some more before you do…you have time.
  10. These patients are sick AF!
    • The mortality of NOAF who Stay in AF is 40-60%!!!!
    • The mortality of NOAF who go to sinus is 26.1% (vs 18% in those without NOAF)
    • AF is likely a marker of illness as opposed to the cause of it so let the AF serve as a marker of someone who needs to stay on your radar!


Antal, O, E Ştefănescu, and N Hagău (2019), ‘Does norepinephrine infusion dose influence the femoral-to-radial mean arterial blood pressure gradient in patients with sepsis and septic shock’, Blood Press Monit, 24 (2), 74-77.

Arrigo, M, D Bettex, and A Rudiger (2014), ‘Management of atrial fibrillation in critically ill patients.’, Crit Care Res Pract, 2014 840615.

Balik, M, I Kolnikova, M Maly, P Waldauf, G Tavazzi, and J Kristof (2017), ‘Propafenone for supraventricular arrhythmias in septic shock-Comparison to amiodarone and metoprolol.’, J Crit Care, 41 16-23.

Barnes, H, G Gurry, D McGiffin, G Westall, K Levin, M Paraskeva, H Whitford, T Williams, and G Snell (2019), ‘Atrial Flutter and Fibrillation Following Lung Transplantation: Incidence, Associations and a Suggested Therapeutic Algorithm.’, Heart Lung Circ

Blecher, GE, IG Stiell, BH Rowe, E Lang, RJ Brison, JJ Perry, CM Clement, B Borgundvaag, T Langhan, K Magee, R Stenstrom, D Birnie, and GA Wells (2012), ‘Use of rate control medication before cardioversion of recent-onset atrial fibrillation or flutter in the emergency department is associated with reduced success rates.’, CJEM, 14 (3), 169-77.

Bosch, NA, J Cimini, and AJ Walkey (2018), ‘Atrial Fibrillation in the ICU.’, Chest, 154 (6), 1424-34.

Brown, SM, SJ Beesley, MJ Lanspa, CK Grissom, EL Wilson, SM Parikh, T Sarge, D Talmor, V Banner-Goodspeed, V Novack, BT Thompson, S Shahul, and to Control Adrenergic Storm in Septic Shock-ROLL-IN (ECASSS-R) study Esmolol (2018), ‘Esmolol infusion in patients with septic shock and tachycardia: a prospective, single-arm, feasibility study.’, Pilot Feasibility Stud, 4 132.

Chapman, MJ, JL Moran, MS O’Fathartaigh, AR Peisach, and DN Cunningham (1993), ‘Management of atrial tachyarrhythmias in the critically ill: a comparison of intravenous procainamide and amiodarone.’, Intensive Care Med, 19 (1), 48-52.

Chow, MS (1996), ‘Intravenous amiodarone: pharmacology, pharmacokinetics, and clinical use.’, Ann Pharmacother, 30 (6), 637-43.

De Backer, D, P Biston, J Devriendt, C Madl, D Chochrad, C Aldecoa, A Brasseur, P Defrance, P Gottignies, JL Vincent, and II Investigators SOAP (2010), ‘Comparison of dopamine and norepinephrine in the treatment of shock.’, N Engl J Med, 362 (9), 779-89.

Delle Karth, G, A Geppert, T Neunteufl, U Priglinger, M Haumer, M Gschwandtner, P Siostrzonek, and G Heinz (2001), ‘Amiodarone versus diltiazem for rate control in critically ill patients with atrial tachyarrhythmias.’, Crit Care Med, 29 (6), 1149-53.

Desai, AD, S Chun, and RJ Sung (1997), ‘The role of intravenous amiodarone in the management of cardiac arrhythmias.’, Ann Intern Med, 127 (4), 294-303.

Faniel, R and P Schoenfeld (1983), ‘Efficacy of i.v. amiodarone in converting rapid atrial fibrillation and flutter to sinus rhythm in intensive care patients.’, Eur Heart J, 4 (3), 180-85.

Fernando, SM, R Mathew, B Hibbert, B Rochwerg, L Munshi, AJ Walkey, MH Møller, T Simard, P Di Santo, FD Ramirez, P Tanuseputro, and K Kyeremanteng (2020), ‘New-onset atrial fibrillation and associated outcomes and resource use among critically ill adults-a multicenter retrospective cohort study.’, Crit Care, 24 (1), 15.

Goldschlager, N, AE Epstein, G Naccarelli, B Olshansky, and B Singh (2000), ‘Practical guidelines for clinicians who treat patients with amiodarone. Practice Guidelines Subcommittee, North American Society of Pacing and Electrophysiology.’, Arch Intern Med, 160 (12), 1741-48.

Goldschlager, N, AE Epstein, GV Naccarelli, B Olshansky, B Singh, HR Collard, E Murphy, and North American Society of Pacing and Electrophysiology (HRS Practice Guidelines Sub-committee (2007), ‘A practical guide for clinicians who treat patients with amiodarone: 2007.’, Heart Rhythm, 4 (9), 1250-59.

Henyan, NN, EL Gillespie, CM White, J Kluger, and CI Coleman (2005), ‘Impact of intravenous magnesium on post-cardiothoracic surgery atrial fibrillation and length of hospital stay: a meta-analysis.’, Ann Thorac Surg, 80 (6), 2402-6.

Hofmann, R, C Steinwender, J Kammler, A Kypta, G Wimmer, and F Leisch (2004), ‘Intravenous amiodarone bolus for treatment of atrial fibrillation in patients with advanced congestive heart failure or cardiogenic shock.’, Wien Klin Wochenschr, 116 (21-22), 744-49.

Hofmann, R, C Steinwender, J Kammler, A Kypta, and F Leisch (2006), ‘Effects of a high dose intravenous bolus amiodarone in patients with atrial fibrillation and a rapid ventricular rate.’, Int J Cardiol, 110 (1), 27-32.

Jardin, F, T Fourme, B Page, Y Loubières, A Vieillard-Baron, A Beauchet, and JP Bourdarias (1999), ‘Persistent preload defect in severe sepsis despite fluid loading: A longitudinal echocardiographic study in patients with septic shock.’, Chest, 116 (5), 1354-59.

Kanji, S, DR Williamson, BM Yaghchi, M Albert, L McIntyre, and Critical Care Trials Group Canadian (2012), ‘Epidemiology and management of atrial fibrillation in medical and noncardiac surgical adult intensive care unit patients.’, J Crit Care, 27 (3), 326.e1-8.

Kim, WY, JH Jun, JW Huh, SB Hong, CM Lim, and Y Koh (2013), ‘Radial to femoral arterial blood pressure differences in septic shock patients receiving high-dose norepinephrine therapy.’, Shock, 40 (6), 527-31.

Kirchhof, P, L Eckardt, P Loh, K Weber, RJ Fischer, KH Seidl, D Böcker, G Breithardt, W Haverkamp, and M Borggrefe (2002), ‘Anterior-posterior versus anterior-lateral electrode positions for external cardioversion of atrial fibrillation: a randomised trial.’, Lancet, 360 (9342), 1275-79.

Kosinski, EJ, JB Albin, E Young, SM Lewis, and OS LeLand (1984), ‘Hemodynamic effects of intravenous amiodarone.’, J Am Coll Cardiol, 4 (3), 565-70.

Lamontagne, F, MO Meade, PC Hébert, P Asfar, F Lauzier, AJE Seely, AG Day, S Mehta, J Muscedere, SM Bagshaw, ND Ferguson, DJ Cook, S Kanji, AF Turgeon, MS Herridge, S Subramanian, J Lacroix, NKJ Adhikari, DC Scales, A Fox-Robichaud, Y Skrobik, RP Whitlock, RS Green, KKY Koo, T Tanguay, S Magder, DK Heyland, and Critical Care Trials Group. Canadian (2016), ‘Higher versus lower blood pressure targets for vasopressor therapy in shock: a multicentre pilot randomized controlled trial.’, Intensive Care Med, 42 (4), 542-50.

Lamontagne, F, A Richards-Belle, K Thomas, DA Harrison, MZ Sadique, RD Grieve, J Camsooksai, R Darnell, AC Gordon, D Henry, N Hudson, AJ Mason, M Saull, C Whitman, JD Young, KM Rowan, PR Mouncey, and trial investigators 65 (2020), ‘Effect of Reduced Exposure to Vasopressors on 90-Day Mortality in Older Critically Ill Patients With Vasodilatory Hypotension: A Randomized Clinical Trial.’, JAMA

Lancaster, TS, MR Schill, JW Greenberg, MR Moon, RB Schuessler, RJ Damiano, and SJ Melby (2016), ‘Potassium and Magnesium Supplementation Do Not Protect Against Atrial Fibrillation After Cardiac Operation: A Time-Matched Analysis.’, Ann Thorac Surg, 102 (4), 1181-88.

Liu, WC, WY Lin, CS Lin, HB Huang, TC Lin, SM Cheng, SP Yang, JC Lin, and WS Lin (2016), ‘Prognostic impact of restored sinus rhythm in patients with sepsis and new-onset atrial fibrillation.’, Crit Care, 20 (1), 373.

Mayr, A, N Ritsch, H Knotzer, M Dünser, W Schobersberger, H Ulmer, N Mutz, and W Hasibeder (2003), ‘Effectiveness of direct-current cardioversion for treatment of supraventricular tachyarrhythmias, in particular atrial fibrillation, in surgical intensive care patients.’, Crit Care Med, 31 (2), 401-5.

McIntyre, WF, KJ Um, W Alhazzani, AP Lengyel, L Hajjar, AC Gordon, F Lamontagne, JS Healey, RP Whitlock, and EP Belley-Côté (2018), ‘Association of Vasopressin Plus Catecholamine Vasopressors vs Catecholamines Alone With Atrial Fibrillation in Patients With Distributive Shock: A Systematic Review and Meta-analysis.’, JAMA, 319 (18), 1889-900.

Morelli, A, C Ertmer, M Westphal, S Rehberg, T Kampmeier, S Ligges, A Orecchioni, A D’Egidio, F D’Ippoliti, C Raffone, M Venditti, F Guarracino, M Girardis, L Tritapepe, P Pietropaoli, A Mebazaa, and M Singer (2013), ‘Effect of heart rate control with esmolol on hemodynamic and clinical outcomes in patients with septic shock: a randomized clinical trial.’, JAMA, 310 (16), 1683-91.

Moss, TJ, JF Calland, KB Enfield, DC Gomez-Manjarres, C Ruminski, JP DiMarco, DE Lake, and JR Moorman (2017), ‘New-Onset Atrial Fibrillation in the Critically Ill.’, Crit Care Med, 45 (5), 790-97.

Rajagopalan, B, Z Shah, D Narasimha, A Bhatia, CH Kim, DF Switzer, GH Gudleski, and AB Curtis (2016), ‘Efficacy of Intravenous Magnesium in Facilitating Cardioversion of Atrial Fibrillation.’, Circ Arrhythm Electrophysiol, 9 

Scheuermeyer, FX, R Pourvali, BH Rowe, E Grafstein, C Heslop, J MacPhee, L McGrath, J Ward, B Heilbron, and J Christenson (2015), ‘Emergency Department Patients With Atrial Fibrillation or Flutter and an Acute Underlying Medical Illness May Not Benefit From Attempts to Control Rate or Rhythm.’, Ann Emerg Med, 65 (5), 511-522.e2.

Sibley, S and J Muscedere (2015), ‘New-onset atrial fibrillation in critically ill patients.’, Can Respir J, 22 (3), 179-82.

Sleeswijk, ME, JE Tulleken, T Van Noord, JH Meertens, JJ Ligtenberg, and JG Zijlstra (2008), ‘Efficacy of magnesium-amiodarone step-up scheme in critically ill patients with new-onset atrial fibrillation: a prospective observational study.’, J Intensive Care Med, 23 (1), 61-66.

Stiell, IG, MLA Sivilotti, M Taljaard, D Birnie, A Vadeboncoeur, CM Hohl, AD McRae, BH Rowe, RJ Brison, V Thiruganasambandamoorthy, L Macle, B Borgundvaag, J Morris, E Mercier, CM Clement, J Brinkhurst, C Sheehan, E Brown, MJ Nemnom, GA Wells, and JJ Perry (2020), ‘Electrical versus pharmacological cardioversion for emergency department patients with acute atrial fibrillation (RAFF2): a partial factorial randomised trial.’, Lancet, 395 (10221), 339-49.

Walkey, AJ, RS Wiener, JM Ghobrial, LH Curtis, and EJ Benjamin (2011), ‘Incident stroke and mortality associated with new-onset atrial fibrillation in patients hospitalized with severe sepsis.’, JAMA, 306 (20), 2248-54.

Walkey, AJ, BG Hammill, LH Curtis, and EJ Benjamin (2014), ‘Long-term outcomes following development of new-onset atrial fibrillation during sepsis.’, Chest, 146 (5), 1187-95.

Walkey, AJ, EK Quinn, MR Winter, DD McManus, and EJ Benjamin (2016a), ‘Practice Patterns and Outcomes Associated With Use of Anticoagulation Among Patients With Atrial Fibrillation During Sepsis.’, JAMA Cardiol, 1 (6), 682-90.

Walkey, AJ, SR Evans, MR Winter, and EJ Benjamin (2016b), ‘Practice Patterns and Outcomes of Treatments for Atrial Fibrillation During Sepsis: A Propensity-Matched Cohort Study.’, Chest, 149 (1), 74-83.

Zhang, B, X Li, D Shen, Y Zhen, A Tao, and G Zhang (2014), ‘Anterior-posterior versus anterior-lateral electrode position for external electrical cardioversion of atrial fibrillation: a meta-analysis of randomized controlled trials.’, Arch Cardiovasc Dis, 107 (5), 280-90.

[a] This Trial is a classic and a must read!

[b] As we all know there are at least 3 definitions of sepsis over time so the definition of this has changed thus shifting our ability to interpret these studies!

[c] This guy is everywhere in AF and Sepsis! Just check out the references!

[d] Look at those waveforms to make sure they have a dicrotic notch!

[e] No association with Kirchoffs rule of electrical current…and they say the universe has no sense of humor! As a refresher this law states that, for any node (junction) in an electrical circuit, the sum of currents flowing into that node is equal to the sum of currents flowing out of that node

[f] Magnesium: I like to speak with my local pharmacist and see what if any policy there is on mag administration because for some reason the OB floor can give a monitored pt 8g of mag but when I do it there is widespread panic. Still this is not the hill to die on though.

[g] Amiodarone can cause phlebitis and used to require a central line but this is no longer true. 

[h] You know who you are!

[i] Bombastus: Paracelsus born Theophrastus von Hohenheim (full name Philippus Aureolus Theophrastus Bombastus von Hohenheim) (1494 –1541). The father of toxicology! Sola dosis facit venenum “Only the dose makes the poison”

[j] Magnesium: Again know your local practices and staff comfort level. However, this is not a hill to die on.






  1. An increase by 4-6-mmol/L [Na] is sufficient to reverse most serious manifestations of acute hyponatremia.
  2. Increase [Na] no more than 10 mEq/L in 24 hour period in pts with LOW RISK Osmotic Demyelination syndrome
  3. Increase [Na] no more than 8 mEq/L in 24 hour period in pts with HIGH RISK Osmotic Demyelination syndrome (ODS).
  4. PTS at HIGH RISK of ODS (mnemonic CHAMP):
    1. Cirhosis
    2. Hypokalemia
    3. Alcoholism
    4. Malnutrition
    5. Plasma Serum sodium <105


  1. For our purposes the only cause of cerebral edema and neurological causes of hyponatremia are the hypotonic (low osmolar) hyponatremia causes.
  2. IF neurological symptoms are present and hypotonic hyponatremia is presumed or confirmed then 150 ml of 3% (3ml/kg) over 20 min
  3. Check serum sodium after 20 min while repeating an infusion of 150 ml of 3% over 20 min IF symptoms persist.
  4. GOAL: Target goal of 5 mmol/L increase in serum Na is achieved


  1. As long as the patient is not hypovolemic AND the SEVERE symptoms are treated then
  2. Use the smallest volume of NS until specific causes are found
  3. Limit the increase in Serum Na to a total of 10 mEq/L[g]in the first 24hr and an additional 8 mEq/L during every 24hr there after until serum sodium reaches a goal of 130 mmol.
  4. Check Serum sodium concentration after 6 hour and after 12 hr and then daily until Na has stabilized.
  5. A sudden increase in urine output to >100ml/h signals increased risk of overly rapid rise in serum sodium concentration


To calculate the anticipated increase of Serum sodium by infused saline:

Screen Shot 2018-12-24 at 5.51.42 AMWhere:

  • Infused Na = 154 if Normal Saline or 513 if 3% hypertonic saline
  • Total Body water (TBW) = Wt x 0.6 if male
  • Total Body water (TBW) = Wt x 0.5 if female
  • Total Body water (TBW) = Wt x 0.5 if elderly (>65) male
  • Total Body water (TBW) = Wt x 0.45 if elderly (>65) female

Example: 70 kg young female with severe cerebral symptoms is given a 3 ml/kg bolus of 3% NaCl (Na=
513 mEq/L) and Serum Na = 108

Infuse Na                   = 513 mEq/L (hypertonic)

Volume infused         = 0.003 L/kg x 70 kg = 0.14 L

Serum Na                   = 108

TBW                            = 70×0.5 (young female) = 35L

Screen Shot 2018-12-24 at 6.33.33 AM

 Final serum Na = 111


  1. BEER POTOMANIA 15 (BOX 1 from algorithm)
    1. NPO except medications for 24 h
    2. No IVFs unless symptomatic
    3. Prescribe IVFs in finite amounts if needed
    4. Serum sodium every 2 h

    5. S Na increase < 10 mEq/L in first 24 h
    6. S Na increase < 18 mEq/L in first 48 h
    7. Re-lower serum sodium levels if necessary
    8. Give all IV medications in D5W

    9. If caloric intake is needed, use D5W.
  2. HYPERVOLEMIC causes (BOX 2 from algorithm)
    1. Recommend against treatment with the sole purpose of aim of increasing the serum sodium concentration in mild or moderate hyponatraemia.
    2. Fluid restriction (<1.5-1.0L/d or 500ml/d less than the Urine output) to prevent further fluid overload.
    3. Recommend against vasopressin receptor antagonists.
    4. Recommend against demeclocycline.
  3. HYPOVOLEMIC (BOX 3/4 from algorithm)
    1. It is recommended to restore volume with i.v. 0.9% saline or a balanced crystalloid solution at 0.5–1.0 ml/kg per h
    2. In case of hemodynamic instability, the need for rapid fluid resuscitation overrides the risk of an overly rapid increase in serum sodium concentration
    3. In these patients restoring volume will suppress vasopressin secretion causing electrolyte-free water excretion to increase. Therefore, these patients are at high risk of an overly rapid increase in serum sodium concentration. Sudden increases in urine output can act as a warning signal that overly rapid correction of hyponatremia is imminent.
  4. EUVOLEMIA HYPOTHYROIDISM (BOX 5 from algorithm)
    1. Unless severe (i.e. myexedma or TSH >50 mIU/mL), other causes of hyponatremia should be sought rather than hypothyroidism. 

    2. Unless the patient has hyponatremic encephalopathy, primary treatment of hyponatremia should consist of thyroid hormone replacement at standard weight-based doses; several days may be needed to normalize the serum Na. 

  5. SIADH (BOX 6 from algorithm)
    1. In moderate or profound hyponatraemia, restricting fluid intake should be first-line treatment.
    2. In moderate or profound hyponatraemia, the following are equal be second-line treatments: increasing solute intake with 0.25–0.50 g/kg per day of urea or a combination of low-dose loop diuretics and oral sodium chloride.
    3. In moderate or profound hyponatraemia, we recommend against lithium or demeclocycline.
    4. In moderate or severe hyponatraemia, we do not recommend vasopressin receptor antagonists.
  6. Cerebral Salt Wasting (Box 7 from algorithm)16 17
    1. CSW is a volume-depleted state
    2. Depending on the severity isotonic or hypertonic solutions are indicated.
    3. Additionally, sodium tablets up to 12 g/d may be used can be combined with the IVF
    4. Once euvolemia is achieved, the goal of therapy is to prevent volume depletion by matching the urinary output.
    5. Fludrocortisone has also been used for the treatment of CSW at doses of 0.1 to 1 mg/d (stimulating reabsorption of sodium and water)
    6. The most common adverse effect associated with fludrocortisone was hypokalemia in up to 73% of patients.


  1. Fluid restriction is first line treatment
  2. Urea[h]/loop diuretics are equal second line treatments


  1. Prompt intervention for re-lowering the serum sodium concentration if it increases >10 mmol/l during the first 24 h or >8 mmol/l in any 24 h thereafter
  2. We recommend discontinuing the ongoing active treatment.
  3. Consulting an expert to discuss if it is appropriate to start an infusion of 10 ml/kg body weight of electrolyte-free water (e.g. glucose solutions) over 1 h under strict monitoring of urine output and fluid balance.
  4. We recommend consulting an expert to discuss if it is appropriate to add i.v. desmopressin 2 mg, with the understanding that this should not be repeated more frequently than every 8 h.


Chances are if you are reading this you have probably seen a case or two of hyponatremia. Well, defining hyponatremia isn’t really hard. In fact treating acute symptomatic hyponatremia isn’t that hard either. So what makes this electrolyte disturbance so painful? It’s the over abundance of math, renal physiology and lack of organization that exists within the literature; enough to make doctors have flashbacks of their USMLE days. Fear not friends, we are going to make this simple and painless. Unless simple and painless are not your thing, then I have for you the sadist’s ball gag equivalent of math in the optional section to appease even the most vicious of IM attendings. No more google-ing 40 different articles to treat that hyponatremic patient, everything in one neat package! So lets get to it!

            First off some definitions:


Classification Definition Comment
Hyponatremia <136  
Moderate 125-135  
Severe/Profound <125 Profound in UK
Acute vs Chronic Time of development (48 hr cutoff)


Usually unknown
Tonicity Dealing with osmolarity There can be unaccounted osmoles (EtOH)
Hypotonic Hypo-osmolar Type of hyponatremia
Volemia Dealing with the pt volume status Difficult to asses clinically.

Table 1. Classification of Hyponatremia (US units: mEq/L) 1, 2

So far, not horrible; Right?


We need to talk about epidemiology not because every chapter on any disease starts out this way but because it will help us down the road. So lets push on. Hyponatremia is probably the most common electrolyte abnormality3.  Based on studies from the first half of the 2000’s the incidence[a]of hyponatremia is about 30-40% of admitted patients worldwide. The largest amount were found in the ICU with hyponatremia <125 mEq being an independent predictor of mortality. Values <126 mEq/L and < 116 mEq/L were found in about 6% and 1% of the patients, respectively4. In ED populations it is obviously lower than in admitted patients. ED based studies outside the US found a prevalence of  3-5% for all adults but 10-17% in those patients age 65 and older. These studies also found an increase in hyponatremia for both groups during the summer months 1-2% greater than the above numbers 5, 6. In the geriatric population 78% of episodes of hyponatremia were precipitated by increased fluid intake, administered orally or as intravenous hypotonic fluid. Finally, up to 17% of chronic alcoholic patients had hyponatremia secondary to Beer Potomania[b]

OPTIONAL: OsmolaLity or OsmolaRity?

To jump back a bit we should define osmolaLity and osmolaRity. Osmolality (with an “L”) is a measure of the osmoles per kilogram (Osm/Kg),  osmolarity (with an “R”) is defined as the number of osmoles per liter (L) (Osm/L). Don’t worry we don’t have to know how many kg our patients are to calculate their osmolality. If the L and the R were flipped it would probably be easier to remember but lets just remember that the L is for weight and the R is for volume. This is important because we calculate osmolaRity but measure osmolaLity so we should keep track of the units. If you want to not be accurate then you don’t have to worry because we are dealing with water in the body, so we can approximate osmolaLity to equal the osmolaRity[c].


Figuring out the differential our patients have for their hyponatremia is probably the most difficult part of this diagnosis. However, here is where I am going to try (or at least try to try) to simplify it.

Click Here for the Hyponatremia Algorithm: Hyponatremia 2.0-3

In order to decide the reason for the hyponatremia in our patients we need to have a starting point. We used to talk about volume status as the starting point but this makes things too difficult furthermore the reliability of volume status by clinical exam is less than accurate. Instead we are going to start with the osmolality.



The first branch point into our differential diagnosis is figuring out what is the serum osmolality (tonicity). We can do this two ways: order an extra lab value of serum osmoles (not usually part of “first round of orders”) or we can estimate (oh thank god!) a serum olsmolarity. To do this we us the formula:

Serum Osm [mmol/L]= 2[Na+] + Glucose + Bun               (Equation 1)

This probably looks a bit strange to US doctors since it’s missing some numbers. That’s because we don’t use [mmol/L] in the US, instead we use the crazy units of  [mEq/L]. So to go from mmol to mEq (or mg same thing for our purposes) we have to divide the molecules by their molecular weight. In fact, any good British child will tell you to convert their glucose to US values by dividing by 18![d]So now our formula becomes the familiar:

Serum Osm [mEq/L]= 2[Na+] + Glucose/18 + Bun/2.8    (Equation 2)


To estimate the serum osmolarity [mEq/L] use the formula:

2[Na+] + Glucose/18 + Bun/2.8


How good is this formula for estimating serum osmoles? When equation 1 (also referred to as the Smithline-Gardner formula) is compared to the mean measured osmolality in healthy patients it gives an osmolal gap (OG, i.e the difference between calculated and measured Osm) close to zero and an SD of 4 mmol/L. When applied to patients in which an osmolality was clinically indicated (i.e where it would theoretically matter) the variability in the OG is approximately +/- 7 mmol/L. This held true as long as ethanol was not adulterating the sample7. Remember that ethanol will contribute to serum osmolarity and will need to be divided by 3.78. Thus equation 2 (in the US) is proposed to be adopted by all clinicians and laboratories along with a “fudge” factor of +/- 10 mmol/L for the gap7. To really get the most accurate answer here you would likely have to order a serum osmoles but it would appear that this estimation holds up pretty well to start treatment.


Only true (hypotonic) hyponatremia matters so exclude other causes.

The normal human plasma osmolality is 275-300 mOsm/Kg9. Therefore, we need to see if we have hypertonic or isotonic hyponatremia. Hypertonic (also called Translocational or redistributive hyponatremia) causes include: hyperglycemia[e], mannitol, and sometimes a contrast load.  Isotonic (also called Pseudo or normo-osmolal hyponatremia) causes include: hyperlipidemia and hyperproteinemia. Once we have excluded these causes by a value of <275 mOsm/kg then we can say that we have a true hypotonic or hypo-osmolar hyponatremia.


Calculate the serum osmolarity. If it is <275 mOsm/Kg then the causes are “true” or hypotonic causes1which can truly cause symptomatic hyponatremia.


The normal response to hyponatremia is marked suppression of ADH secretion, resulting in the excretion of maximally dilute urine with an osmolality below 100 mOsm/kg and a specific gravity ≤1.003. Vasopressin (ADH) levels are what really differentiates the type of hyponatremia at this point. The first type will have high vasopressin levels and high urine Osm with low Na.  Here high vasopressin causes inability to excrete water and consists of the classic 3 groups of hyponatremia: hypovolemic, euvolemic and hypervolemic. The second type will have low serum vasopressin levels and typically low urine osmolality. Thus, mechanisms other than vasopressin are responsible. These types includes: chronic renal failure, psychiatric disorders, potomania, and low solute load excreted in the urine. However we can’t measure Vasopressin (ADH) so we use low osm and low sodium to represent this group. Values above 100 mOsm/kg indicate an inability to normally excrete free water, most commonly because of persistent secretion of ADH. Therefor the urine osmolality can help us. Once again we have 2 options to determine osmolality: order it or to estimate it. 


To jump back a bit to med school physiology lets understand what ADH does. ADH (Vasopressin) is released from the posterior pituitary. It does this in response to 2 motivating factors: 1. Low Blood pressure via Angiotensin II released from the kidney and activating the V2 receptor; 2. High plasma osmolality (Hypertonicity) either by too little water or too many solutes. Once ADH is release it works on the principal cells in the collecting duct to stimulate aqauporin-2 channels to be inserted on the collecting duct to reabsorb more water back into the blood. So ADH “ADDs H20″ to the plasma. Thus in the algorithm when there is low URINE osmolality this must mean that ADH is not working and is “ADH independent”


The normal response to hyponatremia causes a urine osmolality below 100 mOsm/kg. To estimate the urine Osm from the urine we can use the Urine specific gravity (USG) and the following equation:

            UOsm = (USG -1) x 25,000                         Equation 311.


So if the urine Osm are calculated and less than 100 mOsm/kg then it is likely that a relative excess water intake is the umbrella cause of the hypotonic hyponatremia1. Thus the differential includes: Primary polydipsia, beer potomania, water dilution in formulas of infants, tap water enemas in infants, or very low sodium intake. Again this is because the normal response to hyponatremia is suppression of ADH, resulting in the excretion of a very dilute urine If that’s the case you are done and all you need to do is fluid restrict! We will speak more about beer potomania later on since it deserves special attention.

            TAKE HOME POINT: UOsm <100 implies water excess


If urine osmolality > 100 mOsm/kg then likely there is impairment of the renal concentrating ability. Now, it gets a little more complicated. *At this point more lab testing will likely be needed. My feeling is if you are going to order a urine sodium, you might as well order everything at once[f]. At this point I recommend ordering:

Serum:  cortisol, Osmoles, TSH, Uric acid

Urine:  creatinine, potassium, sodium, Uric acid,


In a study by Imran, 504 urine specimens from patients on whom a simultaneously drawn USG and an osmolality were available were examined. They found good linear correlation between USG and Urine Osm when measured either by reagent strip or refractometry. Urine samples were divided into ‘‘clean’’ and ‘‘pathological’’ urines. Pathologic urines on reagent strip included: glucose, ketones, urobilinogen, bilirubin, and protein and ketones, bilirubin, and urobilinogen, for samples measured using refractometry. The study found that pathological urines did not correlate as well to urine Osm as the non-pathological12.  In another study of only hyponatremic patients it was found that USG has a linear relationship with measured urine osmolality in patients with hyponatremia. A multiplying factor of 20-33 is better than 30-40 in predicting urine osmolality of most patients with hyponatremia11. The commonly used formula to predict UOsm from USG uses a multiply of 30 but from the above two studies it would appear that multiplying by 25 gives a closer approximation and less likely to overestimate.  Hence equation 3.

STEP 4: URINE Na <30 mEq/L

If urine sodium concentration ≤ 30 mmol/L, then most likely low effective arterial volume (see below optional section on What the hell does “low effective arterial volume” mean) is the cause of the hypotonic hyponatraemia. Really what we have to focus on here is whether the patient has HYPERvolumic or HYPOvolemic causes. The cutoff of urine sodium of 30 is randomly picked based on studies and is used by guidelines2. Using a cutoff of <30 suggests the above diagnoses and diuretics wont affect this diagnosis1.

OPTIONAL: What the hell does “low effective arterial volume” mean

Early observations in patients with cardiac failure demonstrated renal sodium and water retention that resulted in an increase in extracellular fluid (ECF) volume and edema. This degree of sodium and water retention in normal individuals would lead to an increase in renal sodium and water excretion, yet the reverse occurs in patients with heart failure. A similar sequence of events occurs in patients with cirrhosis and pregnancy. Renal sodium and water retention in edematous disorders continued to be perplexing. A term for sodium and water retention in edematous disorders was proposed to be due to a decrease in “effective blood volume” rather than “total blood volume”. For many years, however, this enigmatic term, which was used to explain sodium and water retention in patients with heart failure or cirrhosis, was never defined.  Use of the term “decreased effective blood volume” can be considered outdated and be replaced by “arterial under-filling.” However, use of the term persists in clinical medicine, as “decreased effective arterial blood volume”13.


Although I said before that determining clinical volume status is more difficult at this stage the causes of hyponatremia from hypovolemic and hypervolemic reasons should be more obvious. Hypervolemic causes would include: CHF, Cirrhosis, and Nephrotic syndrome. Hypovolemic causes would include: Diarrhea and vomiting, third spacing, diuretics. It should be noted that a third cause of low urine sodium is very low sodium intake but this is rare in western diets. Also know there are insufficient data to suggest that increasing serum sodium concentration improves patient-important outcomes in moderate hyponatraemia with expanded extracellular fluid volume, in cirrhosis or heart failure.



If the urine sodium is >30 mEq/L then one needs to consider if the patient is on diuretics. When I say diuretics I’m mostly refering to Thiazide and Thiazide-like diuretics. Potassium sparing diuretics can contribute to hyponatremia but less so. Loop diuretics are much less likely to cause hyponatremia.  In fact diuretics can cause a urine sodium of <30 mEq/L also but we will touch on how to differentiate this as well.  If the patient is NOT on diuretics then again we must decide if the patient has hypovolemia or Euvolemia


Determining volume status in this category may be more subtlebut hypovolemia should be more obvious than euvolemia in this group. Thus, if by clinical exam, the patient is hypovolemic then the causes include: vomiting, primary adrenal insufficiency and renal/cerebral salt wasting.Determining euvolemia is much more subtle, however, if the ECF is normal by clinical exam then the causes include: SIADH, secondary adrenal insufficiency, and hypothyroidism (realistically unless hypothyroidism is severe such as myxedema or TSH >50 mIU/mL, then other causes of hyponatremia should be considered10). 


As I said above the big problem at this branch point is that determining whether a patient is euvolemic or not can be very difficult. Additionally the treatments of SIADH and salt wasting are differentbecause of fluid restricting patients with SIADH as opposed to administering salt and water in salt wasting14. One way to differentiate this is by the fractional excretion of uric acid (urate). FEurate is normally 4%–11%and will stay normal in the excess water states with low urine osmoles because the concentrating ability is preserved (BOX 1 conditions). If the FEurate is <4%, it is consistent with pre-renal (BOX 2,3,4 conditions) including volume depleted states or edematous states such as CHF, cirrhosis, nephrotic syndrome,and pre-eclampsia.


Distinguishing these two can be very different because the only real difference is that in salt wasting there is a decrease in volume. However, the treatments are very different. The FEurate is especially helpful in distinguishing these two entities. In SIADH and Salt wasting (either cerebral or renal) FEurate is increased to >11% while the sodium is low. However, they can be differentiated after correction of the sodiumto >130. In SIADH, correction of hyponatremia will normalize FEurate to <11%, however in salt wasting the FEurate will still be inceased to>11%. One important caviat is that for the FEurate to be valid the patients serum creatinine must be <1.5 mg/dL14.


Essential criteria

  • Effective serum osmolality < 275 mOsm/kg 

  • Urine osmolality > 100 mOsm/kg at some level of decreased effective osmolality

  • Clinical euvolaemia 

  • Urine sodium concentration > 30 mmol/L with normal salt and water intake 

  • Absence of adrenal, thyroid, pituitary or renal insufficiency 

  • No recent use of diuretic agents 

Supplemental criteria

  • Serum uric acid < 0.24 mmol/L (< 4 mg/dL) 

  • Serum urea < 3.6 mmol/L (< 21.6 mg/dL)

  • Failure to correct hyponatraemia after 0.9% saline infusion 

  • Fractional sodium excretion > 0.5% 

  • Fractional urea excretion > 55% 

  • Fractional uric acid excretion > 12%
Correction of hyponatremia through fluid restriction


Fractional Excretion Of Sodium FENa

It is calculated byScreen Shot 2018-12-24 at 5.51.21 AM         

In patients with normal renal function and hyponatremia cut off for FENa is <0.1%.

<0.1%- hypovolemic hyponatremia

>0.1%- hypervolemic and normovolemic hyponatremia.


In hyponatremia due to SIADH, the blood urea nitrogen (BUN) is usually less than 5 mg/dL. However, as urea excretion decreases with aging the absence of a low BUN cannot be used to exclude SIADH in older patients10.

 Fractional Excretion of Uric Acid14

This is calculated as:   Screen Shot 2018-12-24 at 5.51.28 AM     

FEUrate <4% implies: Volume Depletion Addison’s Disease Edematous states, CHF

  Cirrhosis, Nephrotic syndrome

FEUrate 4-11% implies: psychogenic polydipsia

FEUrate >11% implies: HCTZ, Salt Wasting

After normalization of Na to >130 FEurate will be <11% in SIADH and >11% in Salt wasting


Cases of a hypoosmolal (Hypotonic) syndrome in beer drinkers were first described in 1972. Up to 17% of chronic alcoholic patients had hyponatremia.  Although not consistently reported in patients with beer potomania, low urine osmolality on admission laboratory test results was not a consistent finding. In addition to the history of excess beer drinking, often a recent history of binge drinking or illness was present.  This may potentially precipitate a rapid decrease in serum sodium levels. The maximum urinary dilution capability is 50 mOsm/L, a large amount of water (>20 L) must be ingested under normal situations to overwhelm the capacity for urinary dilution. For example, if the patient excretes only 100 mOsm/d, greater than 2 L of fluid intake with a urinary dilution capability of 50 mOsm/L will result in net water retention and subsequently hyponatremia. Patients with beer potomania have a history of significant beer drinking, often long term, in conjunction with a poor diet. The net result is very low osmole intake because beer has very little sodium and no protein, but has some calories that prevent endogenous protein breakdown (urea generation). Because the obligatory solute loss in a day is approximately 250 mOsm in these patients, with a urinary dilution capability of 50 mOsm/L, water intake greater than 5 L (or 14 cans of beer) results in hyponatremia. The net effect is an excess of free water without the solute for diuresis. ADH levels are expected to be suppressed in patients with beer potomania. The low ADH levels limit free-water reuptake in the collecting tubules of the kidney and explain why these patients have brisk diuresis when solute is presented. Sodium chloride in IV fluids is a common source of the solute load while hospitalized. Urine osmolality on recheck after the solute is introduced is low in these patients because of the low ADH levels. Based on a solute concentration of 308 mEq/L (154×2) in 0.9NaCl solution and the kidney’s diluting ability of 50 mOsm/L, significant diuresis can occur with 1 L of NS solution in the setting of a low-ADH state. This water diuresis can produce large increases in serum sodium levels in a short period. Attempting to replace this with electrolyte-free water to prevent a rapid increase in sodium levels can be difficult. Beer potomania is unusual because the cause of hyponatremia is multifactorial, including low osmole intake. Furthermore, as these patients convert to a low ADH state, the rate of correction may be dramatic.  One study found that 18% of patients presenting with beer potomania developed ODS. Three large retrospective reviews of patients who presented with symptomatic severe hyponatremia found no benefit to aggressive correction of chronic hyponatremia. If the patient is asymptomatic, fluid restriction and monitoring the patient despite the degree of hyponatremia is the recommended approach. If the serum sodium level increase occurs at a rate that will exceed the desired goal, D5W infusion should be started to match urine output. The D5W rate can be adjusted every 2 hours based on serum sodium level change. If serum sodium levels increase to greater than either the 24- or 48-hour goals, D5W rate should be increased to decrease the serum sodium level to the recommended goal. Desmopressin may be considered if diuresis occurs at an excessive rate that the infused D5W is unable to match; based upon the current rate of serum sodium level change, the goal will be exceeded despite D5W; the goal has been already been exceeded; or last, symptoms of ODS develop15. A large rise in serum Na after infusion of a test volume of isotonic saline suggests the presence of hypovolemia. 









  1. Spasovski G, Vanholder R, Allolio B et al. Clinical practice guideline on diagnosis and treatment of hyponatraemia. Nephrol Dial Transplant 2014; 29 Suppl 2, i1-i39.
  2. Hoorn EJ, Zietse R. Diagnosis and Treatment of Hyponatremia: Compilation of the Guidelines. J Am Soc Nephrol 2017; 28, 1340-1349.
  3. Anderson RJ, Chung HM, Kluge R, Schrier RW. Hyponatremia: a prospective analysis of its epidemiology and the pathogenetic role of vasopressin. Ann Intern Med 1985; 102, 164-168.
  4. Upadhyay A, Jaber BL, Madias NE. Incidence and prevalence of hyponatremia. Am J Med 2006; 119, S30-5.
  5. Giordano M, Ciarambino T, Castellino P et al. Seasonal variations of hyponatremia in the emergency department: Age-related changes. Am J Emerg Med 2017; 35, 749-752.
  6. Imai N, Osako K, Kaneshiro N, Shibagaki Y. Seasonal prevalence of hyponatremia in the emergency department: impact of age. BMC Emerg Med 2018; 18, 41.
  7. Choy KW, Wijeratne N, Lu ZX, Doery JC. Harmonisation of Osmolal Gap – Can We Use a Common Formula. Clin Biochem Rev 2016; 37, 113-119.
  8. Purssell RA, Pudek M, Brubacher J, Abu-Laban RB. Derivation and validation of a formula to calculate the contribution of ethanol to the osmolal gap. Ann Emerg Med 2001; 38, 653-659.
  9. Plasma osmolality. Wikipedia
  10. Sahay M, Sahay R. Hyponatremia: A practical approach. Indian J Endocrinol Metab 2014; 18, 760-771.
  11. Sumethkula V, Choojitaromb K, Ingsathitc A, Radinahamed P. The Correlation between Urine Specific Gravity and Urine Osmolality in Patients with Hyponatremia. nternational Journal of Sciences: Basic and Applied Research (IJSBAR) 2017; 31, 181-189.
  12. Imran S, Eva G, Christopher S, Flynn E, Henner D. Is specific gravity a good estimate of urine osmolality. J Clin Lab Anal 2010; 24, 426-430.
  13. Schrier RW. Decreased effective blood volume in edematous disorders: what does this mean. J Am Soc Nephrol 2007; 18, 2028-2031.
  14. Maesaka JK, Imbriano L, Mattana J, Gallagher D, Bade N, Sharif S. Differentiating SIADH from Cerebral/Renal Salt Wasting: Failure of the Volume Approach and Need for a New Approach to Hyponatremia. J Clin Med 2014; 3, 1373-1385.
  15. Sanghvi SR, Kellerman PS, Nanovic L. Beer potomania: an unusual cause of hyponatremia at high risk of complications from rapid correction. Am J Kidney Dis 2007; 50, 673-680.
  16. Momi J, Tang CM, Abcar AC, Kujubu DA, Sim JJ. Hyponatremia-what is cerebral salt wasting. Perm J 2010; 14, 62-65.
  17. Yee AH, Burns JD, Wijdicks EF. Cerebral salt wasting: pathophysiology, diagnosis, and treatment. Neurosurg Clin N Am 2010; 21, 339-352.
  18. Filippatos TD, Liamis G, Christopoulou F, Elisaf MS. Ten common pitfalls in the evaluation of patients with hyponatremia. Eur J Intern Med 2016; 29, 22-25.


[a]Incidence Incidence is the rate of new cases of the disease. Prevalence is the actual number of cases alive or the accumulation of the incidences over a period of time. 

[b]Poto – drinking alcohol; mania – excessively

[c]Well really Osmolarity = Osmolality x 0.995 but who is counting.

[d]I was literally once told this by one of my really smart 8 yo Type I diabetic patient from the UK who told me how to convert mEq/L to mmol/L for his glucometer readings!

[e]Remember: each mg increase in blood glucose above 100 mg/dl decreases the serum sodium by 1.6 meq/l. This is negligible when blood sugar is less than 300 mg/dl. When serum triglycerides are above 100 mg/dl, for every 500 mg/dl rise in serum triglycerides, fall in serum sodium will be about 1.0 mEq/L. When serum protein is above 8 gm/dl, for every 1 gm/dl rise in serum protein, fall in serum sodium will be about 4.0 mEq/L10.

[f]This is the Shriki-EM Mantra

[g]US guidelines recommend a limit of 6-8 but this is based on limited data.

[h]As a means of increasing solute intake, daily intake of 0.25–0.50 g/kg urea can be used. The bitter taste can be reduced by the following recipe in Sachets:  10 g urea + 2 g NaHCO3 + 1.5g citric acid + 200 mg sucrose to be dissolved in 50–100 ml water.  Alternately using a commercially available urea powder drink mix (Ure-Na by Nephcentric)2

Stroke 2018 Update

In March New Stroke guidelines came out. Wow are they aggressive! tPA and endovascular therapy (EVT) for all… forget what the literature says! Well We NEED to know these so I’ve listed them here for reference. Sorry about the format. I’ll pretty it up soon!

Stroke. 2018;49:e46–e99. DOI: 10.1161/STR.0000000000000158.)

  1. A primary goal of achieving door-to-needle (DTN) times within 60 minutes in 50% of AIS patients treated with IV alteplase should be established.
  2. It may be reasonable to establish a secondary DTN time goal of achieving DTN times within 45 minutes in 50% of patients with AIS who were treated with IV alteplase.
  3. Systems should be established so that brain imaging studies can be performed within 20 minutes of arrival in the ED in at least 50% of patients who may be candidates for IV alteplase and/or mechanical thrombectomy.
  4. The CT hyperdense MCA sign should not be used as a criterion to withhold IV alteplase from patients who otherwise qualify.
  5. Routine use of magnetic resonance imaging (MRI) to exclude cerebral microbleeds (CMBs) before administration of IV alteplase is not recommended.
  6. Use of imaging criteria to select ischemic stroke patients who awoke with stroke or have unclear time of symptom onset for treatment with IV alteplase is not recommended outside a clinical trial.
  7. For patients who otherwise meet criteria for EVT, it is reasonable to proceed with CTA if indicated in patients with suspected intracranial LVO before obtaining a serum creatinine concentration in patients without a history of renal impairment.
  8. In selected patients with AIS within 6 to 24 hours of last known normal who have LVO in the anterior circulation, obtaining CTP, DW-MRI, or MRI perfusion is recommended to aid in patient selection for mechanical thrombectomy, but only when imaging and other eligibility criteria from RCTs showing bene t are being strictly applied in selecting patients for mechanical thrombectomy.
  9. Only the assessment of blood glucose must precede the initiation of IV alteplase in all patients.
  10. Supplemental oxygen is not recommended in nonhypoxic patients with AIS.
  11. Patients who have elevated BP and are otherwise eligible for treatment with IV alteplase should have their BP carefully lowered so that their systolic BP is <185 mm Hg and their diastolic BP is <110 mm Hg before IV brinolytic therapy is initiated.
  12. Hypoglycemia (blood glucose <60 mg/dL) should be treated in patients with AIS.
  13. IV alteplase (0.9 mg/kg, maximum dose 90 mg over 60 minutes with initial 10% of dose given as bolus over 1 minute) is recommended for selected patients who may be treated within 3 hours of ischemic stroke symptom onset or patient last known well or at baseline state. Physicians should review the criteria outlined in Table 6 to determine patient eligibility.
  14. IV alteplase (0.9 mg/kg, maximum dose 90 mg over 60 minutes with initial 10% of dose given as bolus over 1 minute) is also recommended for selected patients who can be treated within 3 and 4.5 hours of ischemic stroke symptom onset or patient last known well. Physicians should review the criteria outlined in Table 6 determine patient eligibility.
  15. For otherwise eligible patients with mild stroke presenting in the 3- to 4.5-hour window, treatment with IV alteplase may be reasonable. Treatment risks should be weighed against possible bene ts.
  16. In otherwise eligible patients who have had a previously demonstrated small number (1–10) of CMBs on MRI, administration of IV alteplase is reasonable.
  1. In otherwise eligible patients who have had a previously demonstrated high burden of CMBs (>10) on MRI, treatment with IV alteplase may be associated with an increased risk of sICH, and the bene ts of treatment are uncertain. Treatment may be
  2. IV alteplase for adults presenting with an AIS with known sickle cell disease can be bene cial.
  1. IV alteplase should not be administered to patients who have received a treatment dose of low-molecular-weight heparin (LMWH) within the previous 24 hours.
  2. Patients should receive mechanical thrombectomy with a stent retriever if they meet all the following criteria: (1) prestroke mRS score of 0 to 1; (2) causative occlusion of the internal carotid artery or MCA segment 1 (M1); (3) age ≥18 years; (4) NIHSS score of ≥6; (5) ASPECTS of ≥6; and (6) treatment can be initiated (groin puncture) within 6 hours of symptom onset.
  3. Although the bene ts are uncertain, the use of mechanical thrombectomy with stent retrievers may be reasonable for carefully selected patients with AIS in whom treatment can be initiated (groin puncture) within 6 hours of symptom onset and who have causative occlusion of the MCA segment 2 (M2) or MCA segment 3 (M3) portion of the MCAs.
  4. Although the bene ts are uncertain, the use of mechanical thrombectomy with stent retrievers may be reasonable for carefully selected patients with AIS in whom treatment can be initiated (groin puncture) within 6 hours of symptom onset and who have causative occlusion of the anterior cerebral arteries, vertebral arteries, basilar artery, or posterior cerebral arteries.
  5. Although its bene ts are uncertain, the use of mechanical thrombectomy with stent retrievers may be reasonable for patients with AIS in whom treatment can be initiated (groin puncture) within 6 hours of symptom onset and who have prestroke mRS score >1, ASPECTS <6, or NIHSS score <6, and causative occlusion of the internal carotid artery (ICA) or proximal MCA (M1). Additional randomized trial data are needed.
  6. In selected patients with AIS within 6 to 16 hours of last known normal who have LVO in the anterior circulation and meet other DAWN or DEFUSE 3 eligibility criteria, mechanical thrombectomy is recommended. (LEVEL I recommendation)
  7. In selected patients with AIS within 16 to 24 hours of last known normal who have LVO in the anterior circulation and meet other DAWN eligibility criteria, mechanical thrombectomy is reasonable. (Level IIA Recommendation)
  8. Defuse Critereia: Age 18-90 years; NIHSSS ≥ 6; femoral puncture within 6 -16 hours of stroke onset/last known well; pre- morbid mRS2≤2; ICA or M1 occlusion by MRA or CTA AND Target Mismatch Profile on CT perfusion or MRI (ischemic core volume is <70 ml, mismatch ratio is >1.8 and mismatch volume is >15 ml)
  9. Dawn Criteria: Age 18;
failed or contraindicated for IV t-PA; NIHSS ≥10; Pre-stroke -mRS 0-1; Time last seen well to Randomization: 6-24h; <1/3 MCA territory by CT or MRI; ICA- T and/or MCA- M1 occlusion;- Clinical Imaging Mismatch:
A. 80 y/o, NIHSS 10 + core <21 mL B. <80 y/o, NIHSS 10 + core <31 mL C. < 80 y/o, NIHSS 20 + core <51 mL
  10. Administration of aspirin is recommended in patients with
AIS within 24 to 48 hours after onset. For those treated with
IV alteplase, aspirin administration is generally delayed until
24 hours later but might be considered in the presence of concomitant conditions for which such treatment given in the absence of IV alteplase is known to provide substantial bene t or withholding such treatment is known to cause substantial risk.
  11. In patients presenting with minor stroke, treatment for 21 days with dual antiplatelet therapy (aspirin and clopidogrel) begun within 24 hours can be bene cial for early secondary stroke prevention for a period of up to 90 days from symptom onset. The generalizability of this intervention in non-Asian populations remains to be established, and a large phase III multicenter trial in the United States, Canada, Europe, and Australia is ongoing.195
  12. In patients with BP 220/120 mm Hg who did not receive IV alteplase or EVT and have no comorbid conditions requiring acute antihypertensive treatment, the bene t of initiating or reinitiating treatment of hypertension within the rst 48 to 72 hours is uncertain. It might be reasonable to lower BP by 15% during the rst 24 hours after onset of stroke.
  13. Routine placement of indwelling bladder catheters should not be performed because of the associated risk of catheter-associated urinary tract infections.
  14. Use of brief moderate hyperventilation (Pco2 target 30–34
mm Hg) is a reasonable treatment for patients with acute severe neurological decline from brain swelling as a bridge to more de nitive therapy.
  15. Hypothermia or barbiturates in the setting of ischemic cerebral or cerebellar swelling are not recommended.
  16. Because of a lack of evidence of ef cacy and the potential to increase the risk of infectious complications, corticosteroids (in conventional or large doses) should not be administered for the treatment of cerebral edema and increased intracranial pressure complicating ischemic stroke.
  17. Prophylactic use of anti-seizure drugs is not recommended.

ATLS 10th Edition 2018 Update

Since summer is closing and  “Fall” is upon us I thought it would be a good time to “drop” some Trauma knowledge especially since “Crash (2)” is in the ATLS update. Here are the major changes in the 10th Edition of ATLS. Luckily ATLS is getting closer to the evidence and what many of us are already doing. Of course it wouldn’t be an ACS program without their special stamp of approval so they had to change some names around. For example for some reason RSI is now called Drug Assisted intubation. I still think RSI has a better ring to it. Sorry for the format… I”ll update soon!

ATLS 10thEdition changes

Initial Assessment

  • 1 L of fluid
    • Bolus “isotonic” solution 1 L for adults and 20ml/kg for peds <40kg
    • Early blood
  • MTP (1:1:1)
  • TXA
  • Canadian Cspine and nexus


  • RSI is now Drug assisted Intubation
  • VL vs DL

Shock Table

  • Base excess added to shock table
  • Early use of blood
  • Management of Coagulopathic patients
  • TXA
    • TXA over 10 min withing 3 horus of injury
    • 1 g over 8 hour infusion after bolus

Thoracic Trauma

  • Flail chest is out
  • Tracheobronchial injury in
  • Tension
    • Needle
      • 5thICS MAL for adult
      • 2ndICS for child
    • 28-32Fr (vs 36-40)
  • Circulatory arrest algorithm
  • Aortic rupture with bb goal HR <80 MAP 60-70
  • Life threatening injuries during primary survey
    • Airway
      • Obstruction
      • Tracheobronchial tree injury
    • Breathing
      • Tension Pnx
      • Open Pnx
    • Circulation
      • Massive Hmthx (1500)
      • Cardiac Tamponade
      • Traumatic circulatory arrest

Abdominal Trauma

  • No more prostate palpation “no high riding prostate”
  • Flow chart for pelvic fx amended
  • “gentle palpation of the bony pelvis for tenderness”
  • An AP pelvic x-ray may help to establish the source of blood loss in hemodynamically abnormal patients and in patients with pelvic pain or tenderness. An alert, awake patient without pelvic pain or tenderness does not require a pelvic radiograph.

Head Injury

  • Maintain SBP >100 mmHg for adults 50-69
  • Maintain SBP >110 for adults 15-49 or >70
  • Goals of Brain injury
    • Clinical:
      • SBP >100
      • Temp 36-38
    • Monitoring
      • CPP >60 mmhg
      • ICP 5-15
      • PBtO2 >15 mmHg
      • Pulse ox >95%
    • Labs
      • Glucose 80-180
      • Hgb >7
      • INR <1.4
      • Na 135-145
      • PaCO2 35-45
      • pH 7.35-7.45
      • Plt >75000
    • Szr prophyaxis
      • PTS – Postratumatic seizures (wthin in 7 days of injury)
      • Prophylactic use not recommended
      • Phenytoin is recommended todecrease the incidene o faerly posttraumatic szr when benefit is felt to more than harm. Hwevver early PTS has not been associated with worse outcomes


  • Terminology changed to restriction of spinal motion
  • New myotome diagram
  • Canadian Cspine and Nexus

MSK Trauma

  • Wt based IV abx regimen

Thermal injuries

  • 2ml/kgxwtx%burn adults
  • 3ml/kgxwtx%burn children
  • Fluid titrated to UO
  • Flame vs electrical all ages are 4ml/kg LR x%TBSA

Pediatric Trauma

  • Needle thoracentesis unchanged
  • Limiting crystalloid
    • 20 ml/kg bolus followed by blood 10-20 ml/kg rbc or ffp and platelet

Transfer to Definitive Care

  • Specific mention of avoiding CT in primary hospital
    • “Do not peform procedures (DPL or CT) that do not change the plan of care
  • SBAR for communcation




Hyper K, The EBM Way: Protect, Push, and Purge

As we all know hyperkalemia is a life threatening condition. But how can something so basic be shrouded in such confusion? So many choices and everyone has their own recipe to fix it. Along with all the treatment choices, come pitfalls and side effects. Thus, we need to know what is the best way to treat this without giving our patients further problems! Enter “The Algorithm”. I hold before you an approach (evidenced based of course) for treating this life threatening condition. Remember, as always this is only one option and not THE ONLY option for treatment. If your hospital has a guideline or best practice then best stick to that. Otherwise, feel free to enter into a deep forest overgrown with evidence and lush with literature about the treatment of hyperkalemia.


Hyperkalemia can be tricky. We all learn the basic ECG patterns for hyperK as it goes through stages on its way to a “sine wave” and eventual flat line. However, experience teaches that hyperK comes in many forms. The most devious is the severe symptomatic bradycardia. All too often a patient will come in for bradycardia and hypotension, then get started on the path to a pacer, only to find out their K was 10! Don’t let this happen to you! Remember the mantra:

O WATA GOO SIAM… wait no not that one this one:

Bad Brady? Give Gluconate”(Ca+2that is)

Let me repeat that “Bad Brady? Give Gluconate”( Ca+2 that is). That’s right for BAD BRADYcardia GIVE calcium GLUCONATE. Or maybe better:


Whatever it takes for you to remember, check that K when you see symptomatic bradycardia and if you don’t have time for it to result give a trial of calcium gluconate before you put in that pacer!



While we are on the topic of electricity, how good is the ECG for detecting hyperK? It depends what you are calling “ECG Changes”. Studies looking at this topic have shown that hyperK can present without any ECG findings. One study found peaked T-waves in only 34% of ECG’s and potassium levels >6.5 mEq/L1. There are also case reports of patients with potassium levels >10 mEq/L and ECG’s non-diagnostic for hyperkalemia2.  In another study it was not until the serum K > 8 that they found 100% concordance between hyperkalemia and ECG change3. So what are some predictors of hyperkalemia in the ECG? This study3suggests the best predictors by Risk Ratio (RR) for hyperkalemia are QRS prolongation (RR 4.74), junctional rhythm (RR 7.46), and bradycardia less than 50 bpm (RR 12.29) (See Table 1). Importantly, these were the ECG changes that were predictive of adverse events in hyperkalemia. Interestingly peaked T-waves in this study had an RR of 0.77[a]3.


Table 1. ECG changes associated with Hyerkalemia3

ECG Finding Risk Ratio, 95% Confidence interval
Bradycardia (HR<50) 12.3, 95% CI [6.69-22.57]
Junctional rhythm 7.46, 95% CI [5.28-11.13]
QRS prolongation >111 4.74, 95% CI [2.01- 11.15]
Peaked T-waves 0.77, 95% CI [0.35-1.70]



We all have that picture in our mind of who gets hyperkalemia but I want to talk about one group in particular that can be at a special risk. Patients with cardiac devices (pacemakers/AICDs) in addition to the typical cardiovascular collapse can have additional problems. Hyperkalemia can precipitate pacemaker problems specifically widening of the QRS, failure to capture, and delay of the interval from the pacemaker stimulus to the onset of depolarization 4.



OK, So now on to the good stuff! The treatment can be divided essentially into 3 phases, which I have dubbed: PROTECT, PUSH, PURGE (mostly because I enjoy  the alliteration[b]).

Hyper K algorithm

PDF VERSION: Hyper K algorithm-2


The first commandment of treating hyperkalemia is to PROTECT(and don’t covet) this myocardium! Calcium is the mainstay of treatment to protect the cardiac cells from the electrical disaster that is hyperkalemia. Calcium gluconate is typically the first line treatment given the fact that there is less elemental calcium than its partner calcium chloride. This becomes both its advantage and hindrance. The gluconate will not likely sclerose the veins given its concentration. Furthermore, it wont cause tissue necrosis (the opposite of helping) or thrombophlebitis unlike the chloride form. However, it may need to be repeated since it contains a third less the calcium than the chloride form. Strangely enough even in the year 2018, the optimal dose of calcium gluconate isn’t known5, however usually recommended is 10mL or 1g IV. There used to be some lore that the gluconate would require liver activation however both animal and human studies have shown that this is not the case even in liver failure5. Therefore, this is unlikely an issue. Due to the potential for harm, the chloride form is only recommended in the case of cardiac arrest or in the presences of a central line.  Calcium may be re-dosed twice based upon expert consensus5, 6.


The second commandment of the algorithm; thou shalt PUSH thine K into thine cells. By pushing the K back into the cells you are merely hiding the K from the now excited myocardium. Remember, this is only a temporary fix. Potassium is pushed into the cell using a beta-2 agonist such as albuterol. Catecholamines work on the sodium-potassium ATP-ase pump. There is nothing special about albuterol (sorry albuterol), any catecholamine will do. However, albuterol is probably the safest and the easiest to administer. Epinephrine could do the same but would have too many side effects so its not really used. The does of albuterol is big however 10-20mg nebulized. These doses can cause a decrease of serum K+ by 0.6 mEq/L within 30 min and 1.0 mEq/L 1 h after administration for 10mg and 20 mg, respectively5. Tell your patient to expect a few jitters with this and consider that up to 40% of patients on oral beta-blocker therapy may be resistant to this therapy. However, predicting who gets this is impossible. The second drug to accomplish this PUSH is insulin given with glucose to prevent hypoglycemia. Insulin can decrease K+ by 0.6–1.2 mEq/L 5, with an onset of action about 15 min, peak around 30-60 min, and last about 4 hours6.  Insulin/glucose combined with albuterol appear to have a synergistic effect5. The last treatment is controversial. The use of 8.4% Bicarbonate has long been held as the go to treatment. Recently, however, the thrill is gone. There is insufficient data showing benefit for bicarb. Several studies reported that sodium bicarbonate did not lower serum potassium significantly or promptly. In one (poorly done) study 1 mEq (1 amp) of bicarb decreased serum K only by 0.15 mEq/L 5. No study has found immediate reductions in serum K+ and the effects may be not be observed until 4-6 hours later5–7. Sodium bicarbonate may expose patients to a large fluid load, hypernatremia, and metabolic acidosis. Therefore, sodium bicarbonate should no longer be the first line therapy for hyperkalemia6. Bicarb may be useful when patients are acidotic or hypovolemic requiring a fluid load given its large amount of sodium6, 7. Table 2 is a summary of the PUSH therapies.

Table 2. Therapies to PUSH K intracellularly.

Calcium (either) 10 ml Immediate 30 min N/A
Albuterol 10mg 25-30 min 1 hr 0.6
Albuterol 20mg 30 min 2 hr 1.0
Insulin 10-20 mg 15 min 4 hr 0.6-1.2
Bicarbonate 4-6 hrs 4-6hrs 0.15?


Obviously, giving doses of insulin can result in a decrease in blood glucose and even life threatening hypoglycemia. Therefore glucose is given in conjunction with insulin. The question is how much insulin and how much glucose? No one knows the answer to this question. Many different schemas have been suggested. Altering the insulin dose has been suggested; from ultra short acting insulin to infusions of insulin. Additionally, altering the amount of glucose; using a continuous infusion vs. bolus has also been suggested. There is no consensus between expert panels on the dose or route of insulin or glucose. Frequent monitoring for hypoglycemia is definitely recommended. One poorly done before and after study suggested 5 units of insulin in ESRD patients (see my post on that study here). One conservative dosing regimen is given below in table 3.  


TABLE 3. Insulin/Glucose Dosing options

Insulin Options Glucose Dose Glucose Re-Dose Lab Monitoring
Regular 5 U

(Consider in ESRD)

25g D50

(1 amp)

None BS q 1 hr x3

BMP 1 hr post insulin

Regular 10 U 25g D50

(1 amp)

25g D50 @ 1 hr post insulin BS q 30 min x 6

BMP 2hr post insulin

Regular 10 U

(Consider if glucose <100 mg/dL)

D10W gtt @

200 ml/h infusion

None BS q 1hr x3

BMP 2hr post insulin

ESRD = End Stage Renal Disease, BMP = Basic Metabolic Panel, glucose mg/dL


This phase will depend on whether or not the patient has working kidneys. In the case that the kidneys are not working, there is a post obstructive uropathy, or there is oliguria; hemodialysis (HD) is the answer (or CRRT). This means placing a call to your friendly neighborhood nephrologist and dialysis nurse and getting them to come out. HD is also the gold standard treatment. Although there are no definitive studies on the timing or dose of HD. Recommendations for initiation can include: persistent ECG changes, poor response to treatments, and severe AKI. Kayexalate (Sodium PolyStyrene/SPS), once the mainstay of treatment, has fallen out of favor. Multiple consensus panels have endorsed not using it 6, 7.  One such panel recommended “that all other treatment options be exhausted prior to using this [SPS] potentially harmful therapy with little evidence of efficacy”6. Subsequently, the FDA has added a warning for colonic necrosis to the Kayexalate labeling when co-administered with sorbitol8. If, on the other hand, the kidneys are working then loop diuretics may be used at a dose of 20 mg (naive and CKD stage <3) or 40 mg (not naive or CKD stage ≥3).


On the horizon are new potassium binding medications. One expert consensus panel[c]recommended using Patiromir in the acute setting7despite ZERO studies performed in the ED.  Patiromer is an FDA approved new potassium binder that exchanges calcium for potassium for treatment of chronic hyperkalemia.  Side effects include binding oral medications, constipation, and hypomagnesaemia.  Furthermore, its effect is not reached until about 7 hours. Another possibility is the yet unapproved Sodium zirconium cyclosilicate (ZS-9). In a study of 45 patients with serum potassium concentrations of at least 6 mEq/l, 10 g of ZS-9 reduced the serum K by 0.4 mEq/l at 1 hour, by 0.6 mEq/l at 2 hours, and by 0.7 mEq/l at 4 hours9.


One last item to tackle is the “Stone Heart” condition. The “stone heart” theory ascribes calcium as the precipitating condition in which the heart is unable to contract. This occurs when a patient with hyperkalemia is given calcium while currently taking digoxin. The thought is that this may be due to the failure of diastole from calcium binding to troponin and the heart freezes like a stone. This has been “romanticized” into lore. Besides how can one deny a sexy name like “Stone Heart”, it rings of truth! Thus far however, not animal studies, case reports, nor retrospective reviews have found an association of mortality with administration of calcium for hyperkalemia in occult digoxin toxicities 10–12.  The romance may be gone. It was a good run “Stone Heart”. I’ll miss you most of all.


How bad is Hyperkalemia? That question can be seen in the graph below from a study by  Einhorn in archives of internal medicine 2009. This study looked at 66,259 Hyperkalemia events (not patients) in a VA population. They found a 2.4% incidence of death WITHIN ONE DAY (yes one day!). The found that in the patient with no chronic kidney disease a serum K between 5.5 and 6.0 had an OR of death within 24 hours of 10 and above 6.0 had an OR of death of almost 32! Those are huge numbers. In actual cases the group with no CKD had an inpatient mortality of 3.2% for K 5.5 to 6.0 meq/L and 8.6% for those with K greater than 6! For the CKD group (as a whole) it was 1.8% for K between 5.5 and 6.0 and 4.8% for those with a k >6.0. Those numbers while about half of these with chronic kidney disease still represents a very high mortality! Thus Hyperkalemia should be taken very seriously and both treated and admitted unless they are very reliable or have good follow up. Screen Shot 2019-03-19 at 10.29.59 PM



  1. Freeman K, Feldman JA, Mitchell P et al. Effects of presentation and electrocardiogram on time to treatment of hyperkalemia. Acad Emerg Med 2008; 15, 239-249.
  2. Szerlip HM, Weiss J, Singer I. Profound hyperkalemia without electrocardiographic manifestations. Am J Kidney Dis 1986; 7, 461-465.
  3. Durfey N, Lehnhof B, Bergeson A et al. Severe Hyperkalemia: Can the Electrocardiogram Risk Stratify for Short-term Adverse Events. West J Emerg Med 2017; 18, 963-971.
  4. Barold SS, Herweg B. The effect of hyperkalaemia on cardiac rhythm devices. Europace 2014; 16, 467-476.
  5. Long B, Warix JR, Koyfman A. Controversies in Management of Hyperkalemia. J Emerg Med 2018; 55, 192-205.
  6. Rossignol P, Legrand M, Kosiborod M et al. Emergency management of severe hyperkalemia: Guideline for best practice and opportunities for the future. Pharmacol Res 2016; 113, 585-591.
  7. Rafique Z, Weir MR, Onuigbo M et al. Expert Panel Recommendations for the Identification and Management of Hyperkalemia and Role of Patiromer in Patients with Chronic Kidney Disease and Heart Failure. J Manag Care Spec Pharm 2017; 23, S10-S19.
  8. Sterns RH, Rojas M, Bernstein P, Chennupati S. Ion-exchange resins for the treatment of hyperkalemia: are they safe and effective. J Am Soc Nephrol 2010; 21, 733-735.
  9. Sterns RH, Grieff M, Bernstein PL. Treatment of hyperkalemia: something old, something new. Kidney Int 2016; 89, 546-554.
  10. Hack JB, Woody JH, Lewis DE, Brewer K, Meggs WJ. The effect of calcium chloride in treating hyperkalemia due to acute digoxin toxicity in a porcine model. J Toxicol Clin Toxicol 2004; 42, 337-342.
  11. Levine M, Nikkanen H, Pallin DJ. The effects of intravenous calcium in patients with digoxin toxicity. J Emerg Med 2011; 40, 41-46.
  12. Van Deusen SK, Birkhahn RH, Gaeta TJ. Treatment of hyperkalemia in a patient with unrecognized digitalis toxicity. J Toxicol Clin Toxicol 2003; 41, 373-376.


[a]For those of you keeping score you might say with an RR of 0.77 peaked T-waves are PROTECTIVE of adverse events of hyperkalemia but this is probably because we are used to looking for peaked T-waves so physicians were more likely to recognize these and treat earlier rather then it being protective of adverse events in hyperK

[b]I always get confused between alliteration, consonance and assonance. So FYI, the difference is in where the rhyme occurs. Alliteration it’s the beginning (e.g. Larry likes Laurie), consonance it’s the end (frog on a log), and assonance its the middle (e.g. rock in a box was locked)

[c]This panel discussion was funded by Relypsa and facilitated by Magellan Rx Management. Relypsa is the manufacturer of Veltassa (patiromer). Rafique is a principal investigator for Relypsa and serves as a consultant for Instrumentation Laboratory, Magellan Health, Relypsa, and ZS-Pharma. Butler serves as consultant for Amgen, AstraZeneca, Bayer, Boehringer Ingelheim, CardioCell, Janssen, Merck, Novartis, Relypsa, and ZS-Pharma. Lopes and Farnum are employed by Magellan Rx Management. Rafique designed the management protocol for this panel discussion and contributed to the writing and editing of this report document. The other authors report no conflicting interests.





Today’s algorithm is all about Trauma in pregnancy. Its a quick breakdown of the need-to-know basics!




1.     Oxygen supplementation should be given to maintain maternal oxygen saturation > 95% to ensure adequate fetal oxygenation

2.     If needed, a thoracostomy tube should be inserted in an injured pregnant woman 1 or 2 intercostal spaces higher than usual

3.     vasopressors in pregnant women because of their adverse effect on uteroplacental perfusion should be used only for intractable hypotension that is unresponsive to fluid resuscitation

4.     After mid-pregnancy, the gravid uterus should be moved off the inferior vena cava. This may be achieved by manual displacement of the uterus or left lateral tilt.

5.     To avoid rhesus D (Rh) alloimmunization in Rh-negative mothers, O-negative blood should be transfused when needed until cross-matched blood becomes available

6.     A Caesarean section should be performed for viable pregnancies (≥ 23 weeks) no later than 4 minutes following maternal cardiac arrest to aid with maternal resuscitation and fetal salvage


1.     FAST is recommended in pregnancy Sn and Sp similar to non-pregnant females

2.     Radiation exposure with a cumulative dose of > 5 rads (50 mGy) is associated with an increased risk to fetus but limited to < 18 weeks gestation. See Appendix for details

3.     Approximate values for CT radiation for >16 wks with cutoff of 250 mGy recommended:

a.     CT Head/Neck and CXR are negligible

b.     CT chest = 0.66 mGy

c.     CT abd = 35 mGy

d.     CT pelvis = 50 mGy

e.     Total for CT C/A/P = 85.66 mGy


1.     If the fetus is viable (≥ 23 weeks), fetal heart rate auscultation and fetal monitoring can be initiated as soon as feasible.

2.     In cases of vaginal bleeding at or after 23 weeks, speculum or digital vaginal examination should be deferred until placenta previa is excluded by a prior or current ultrasound scan

3.     In addition to the routine blood tests, a pregnant trauma patient should have a coagulation panel including fibrinogen.

4.     KB testing should be done in all Rh- negative pregnant trauma patients.

5.     RhogamTM, anti-D IgG, should be given to all Rh D-negative pregnant trauma patients. In Rh-negative pregnant trauma patients, quantification of maternal–fetal hemorrhage by Kleihauer-Betke should be done to determine the need for additional doses.

a.     A single dose of 300 mcg, administered within 72 hours of injury, provides protection against sensitization for up to 30 mL of fetal blood in the maternal circulation.

b.     The feto-placental blood volume is estimated to be 120 mL/kg of fetal weight.

c.     In most cases of traumatic maternal–fetal hemorrhage, the estimated volume of fetal blood in the maternal circulation is less than 15 mL and in more than 90% of cases it is less than 30 mL.

d.     Therefore, the vast majority of Rh-negative patients are protected by one dose.

e.     If the KB test indicates transplacental hemorrhage in excess of 30 mL fetal blood, additional doses of anti-D IgG may be required.

6.     Tetanus vaccination is safe in pregnancy and should be given when indicated



1.     All pregnant trauma patients with a viable pregnancy (≥ 23 weeks) should undergo electronic fetal monitoring for at least 4 hours. Ctxs <6/hour consider discharge, Ctxs ≥6/hour consider admission

a.     ACOG recommends a minimum of 2-6 hours of monitoring post-trauma

b.     Abruption has been reported to occur up to 24 hours after a traumatic insult.  It has not been reported when <1 contraction is present in any 10-minute interval over a 4-hour period.

c.     Thus, fetal monitoring can be discontinued after 4 hours if uterine contractions occur less frequently than every 10 minutes, the fetal heart tracing is reassuring, and there is no maternal abdominal pain or vaginal bleeding.

2.     Pregnant trauma patients (≥ 23 weeks) should be admitted for 24-hour observation in the setting of:  uterine tenderness, significant abdominal pain, vaginal bleeding, sustained contractions, rupture of the membranes, atypical or abnormal fetal heart rate pattern, high risk mechanism of injury(motorcycle, pedestrian, high speed crash), or serum fibrinogen < 2 g/L

a.     In a prospective cohort study, 85 pregnant women (12 to 41 wks gestation) were compared with a control group of pregnant women matched for gestational age. Study subjects whose placentas were anteriorly placed were at increased risk for fetomaternal transfusion on comparison with other placental positions (47% vs 23.5%, p less than 0.05). Immediate adverse outcomes including abruptio placentae occurred frequently in the study group (9.4%) and were not predictable on the basis of injury severity. Four hours of CTM used as a screening tool was found to be an extremely sensitive (100%) but nonspecific indicator of immediate adverse outcomes. This study recommended that routine screening for fetomaternal transfusion occur in all pregnant women who suffer trauma during pregnancy beyond 11 weeks’ gestation and that a minimum of 4 hours of cardiotocographic monitoring occur in women greater than 20 weeks’ gestation. Patients were discharged home if contractions ceased or were less frequent than once every 15 minutes.  Source: Pearlman MD, Tintinalli JE, Lorenz RP. A prospective controlled study of outcome after trauma during pregnancy. Am J Obstet Gynecol.

b.      Another retrospective study of 271 pregnant patients, suggested monitoring for at least 24 hours only for a selected group of patients at high risk conditions. This high-risk group consisted of patients involved in motorcycle, pedestrian or high velocity collisions, those ejected from motor vehicles and patients demonstrating maternal tachycardia, abnormal fetal heart rate pattern, and high injury severity scores.  Source: Curet MJ, Schermer CR, Demarest GB, Bieneik EJ 3rd, Curet LB. Predictors of outcome in trauma during pregnancy: identification of patients who can be monitored for less than 6 hours. J Trauma 2000;49:18–24.



1.     Every woman who sustains trauma should be questioned specifically about domestic or intimate partner violence.

2.     During prenatal visits, the caregiver should emphasize the importance of wearing seatbelts properly at all times



a.     DOSE measured in Rads or Gray(Gy) [1000mGy = 1Gy]

i.     Amount of energy deposited per kg of tissue

b.     Relative Effective Dose measured in Sieverts

                                               i.     Amount of energy deposited per kg of tissue normalized for              biological effectiveness

                                              ii.     1 Gy ~ 1 Sievert

2.     Effect of Gestation Age on Exposure

a.     0-2 weeks         all or none effect (Death of embryo)       50-100 mGy

b.     2-8 weeks         Birth Defect/IUGR                                        200-250 mGy

c.     8-15 weeks       Intellectual disability (high risk)               60-310 mGy

d.     8-15 weeks       IQ drop                                                           25pts / 1Gy

e.     8-15 weeks       microcephaly                                                200 mGy

f.      16-25 wks         Intellectual disability (low risk)                250-280 mGy

3.     Fetal Radiation with Common Testing

a.     Very low dose <0.1mGy

                                               i.     C-spine x-ray 2view

                                              ii.     Head or neck CT

                                             iii.     Extremity x-ray

                                             iv.     Chest x-ray (two views)

b.     Moderate dose <10mGy

                                               i.     Abdominal x-ray (3mGy)

                                              ii.     Lumbar x-ray (10mGy

                                             iii.     Chest CT or CT PE (0.66mGy)

c.     Higher-dose examinations (10–50 mGy)

                                               i.     Abdominal CT   (35 mGy)

                                              ii.     Pelvic CT          (50 mGy)

Source: ACOG Guidelines for Diagnostic Imaging During Pregnancy and Lactation e210 VOL. 130, NO. 4, OCTOBER 2017. Obstetrics and Gynecology




C. Diff-(iccile): Should the treatment B. Diff-(erent)? A 2018 Update

Table 1. Adult Treatment Strategies 1:

Severity Severity Definition Treatment
First time, non-severe WBC ≤15000 cells/mL and Cr <1.5 mg/dL • vancomycin PO 125 mg QID x 10 days, OR
• fidoxamycin 200 mg BID x 10 days
• Alternate: flagyl, 500 mg PO TIDx10 d
First time, severe WBC ≥15000 cells/mL or Cr >1.5 mg/dL • vancomycin PO 125 mg QID x 10 days, OR
• fidoxamycin 200 mg BID x 10 days
First time, fulminant Hypotension or shock, ileus, megacolon • vancomycin 500 mg QID x 10 days PO or per NGT. If ileus, consider adding rectal instillation of vanc + IV metronidazole (500 mg every 8 hours)
Recurrence • Vancomycin in a tapered and pulsed regimen, OR
• Vancomycin for 10 days followed by rifaximin for 20 days, OR
• Fecal microbiota transplantation
• 125 mg qid
• Vancomycin: 500 mg qid; rifaximin: 400 mg tid

Table 2. Pediatric Treatment Strategies1

Definition  Treatment  Dose 
First time, non-severe • Metronidazole PO ×10d OR
• Vancomycin PO ×10d
• 7.5 mg/kg/dose tid
• 10 mg/kg/dose qid
Initial episode, severe/ fulminant • Vancomycin PO or PR +/-

Metronidazole IV ×10d

• 10 mg/kg/dose qid
• 10 mg/kg/dose tid
First recurrence, non-severe • Metronidazole ×10d PO

• Vancomycin ×10d PO

• 7.5 mg/kg/dose tid or qid
• 10 mg/kg/dose qid

Figure 1. Possible Testing Algorithm Recommendation 5

C.Diff pic

C.Diff testing

Let’s start with a few of the important recommendations on C. Diff Infections (CDI) from IDSA:


 Whom to Test – Adults

Patients with unexplained and new-onset ≥3 unformed stools in 24 hours are the preferred target population for testing for CDI. NEVER test formed stool

Whom to Test – Kids

Because of the high prevalence of asymptomatic carriage of toxigenic C. difficile in infants, testing for CDI should never be routinely recommended in infants ≤12 months of age with diarrhea. Clostridium difficile testing should not be routinely performed in children with diarrhea who are 1–2 years of age unless other infectious or noninfectious causes have been excluded. In children ≥2 years of age, C. difficile testing is recommended for patients with prolonged or worsening diarrhea and risk factors (e.g., underlying inflammatory bowel disease or immunocompromising conditions.

How to Test

Use a stool toxin test as part of a multistep algorithm. There are insufficient data to recommend use of fecal lactoferrin. See Figure 1 below.

How to Isolate

Accommodate patients with CDI in a private room with a dedicated toilet to decrease transmission to other patients. Do not group patients with CDI who have other multidrug-resistant organisms. Patients with suspected CDI should be placed on preemptive contact precautions pending the C. difficile test results. In CDI outbreaks settings, perform hand hygiene with soap and water preferentially instead of alcohol-based hand hygiene products before and after caring for a patient with CDI.  There is increased efficacy of spore removal with soap and water. Handwashing with soap and water is preferred if there is direct contact with feces or an area where fecal contamination is likely.


See Table 1 and 2 below for treatment.


There are insufficient data at this time to recommend probiotics for primary prevention of CDI.


The above are the IDSA guidelines. But how does this affect us in practice? We all often treat patients with limited resources so how will using PO vancomycin as first line affect cost? To that end I offer up these two studies. This first study2 in 2018 of a MATHEMATICAL MODEL of IN-PATIENT data showed that overall percentage of patients cured was: fidaxomicin (96.53%), vancomycin (95.19%) and metronidazole (94.23%). The expected cost of treatment was lowest for vancomycin ($1,306.62), followed by metronidazole ($1,553.01) followed by fidaxomicin ($5,095.70).  Next up is this 2017 Cochrane review3 for efficacy of the same antibiotics. The authors looked at using mostly outpatient cases but few had severe C. diff.  When comparing cure for vancomycin to metronidazole they found 79% (339/428) for vancomycin versus 72% (318/444) for metronidazole. When comparing fidaxomicin to vancomycin they found fidaxomicin more effective for achieving symptomatic cure with 71% (407/572) for fidaxomicin versus 61% (361/592) for vancomycin. Using 2016 costs they found metronidazole cost to be the least expensive at $13, vancomycin to be the next more expensive at $1779, and fidaxomicin to be the most at $3454.83.  Lastly the guidelines say there is insufficient evidence for treatment of probiotics. However, it is probably helpful and not harmful to recommend a probiotic for the prevention of antibiotic associated CDI when prescribing high risk antibiotics like the fluoroquinolones and clindamycin. This Cocharane4 review found a benefit and reduced the risk of adverse events when recommending probiotics along with antibiotics by 17% (RR 0.83, 95% CI 0.71 to 0.97).

Commentary on Treatment

For in-patient treatment I think the data shows pretty well that oral vancomycin is more effective and presents a cost savings. Of course, we have to realize that the first study makes A LOT of assumptions but I think its face value probably makes sense. The big difference here is WETHER OR NOT YOUR HOSPITAL compounds vancomycin from the IV form or gets it de novo.

For out-patient treatment I called a compounding pharmacy and was told that this past January a dedicated oral vancomycin product came out. Therefore, they can no longer compound vancomycin.  So now, if my patient has low resources and “follow up” (i.e. come back to the ED seems to be the default lately) then I would continue to go with metronidazole. You should explain to the patient there is a small chance that the metronidazole won’t work (1-5%, or for every 20 people choosing metronidazole one will fail to be cured; Number needed to fail). However, this beats no treatment at all if you can’t afford the pharmaceutical-company-gouging prices of non-compounded vancomycin. When I called a local pharmaceutical company with stores nationally I was told the cost for a course of vancomycin 125 mg PO WITHOUT insurance was $6100 and the cost for a course of metronidazole was $31.69

Take Home on Treatment

For outpatient ONLY use vancomycin 125 mg PO QID IF the patient has insurance and maybe GIVE THEM A BACK UP Rx FOR METRONIDAZOLE otherwise metronidazole may still be the best bet with a small but significant treatment failure rate. Recommend a probiotic when prescribing antibiotics with a high risk of C. Diff.

For inpatient use vancomycin compounded and this should be efficacious and cost effective.


Commentary on Testing:

There are a LOT of tests and even more acronyms when it comes to C. Diff testing. I’m going to try to simplify them. First there are the NAATs. These are the PCR tests or nucleic acid amplification tests. These are highly sensitive. Almost too sensitive as the can detect C. Diff even after a patient has been clinically cured. So, they are great for first timers but not if you have had C. Diff in the past. Then there are the toxin tests. These detect free toxins in stools and are therefore believed to correlate to clinical symptoms plus they are cheap and easy. Don’t celebrate yet because, sensitivity of Toxin A/B immunoassay (EIA) is suboptimal. Lastly, there are the GDH immunoassays. They are also easy to perform and cheap. They detect glutamate dehydrogenase (GDH), an enzyme that is produced by both toxigenic and non-toxigenic strains and therefore also may not correlate to clinical symptoms.  This is why, algorithms are usually recommended. Even then it might still find only colonization and not pathogenic bacteria. Luckily for us there is a definitive treatment: STOOL TRANSPLANT! Where do I sign up to be a donor? The best thing to do (as with any test) is find out what your specific lab does for algorithms and what are the weaknesses. A common algorithm is below. Sometimes a fecal lactoferrin can help distinguish active disease from non-active disease so I added it in but it’s not really part of the algorithm.

Take Home on testing:

Talk to your lab and find out what the recommended algorithm is. Never test formed stool. Use some combination of PCR and Toxin seems to be best. Consider adding fecal lactoferrin. 





  1. McDonald, L. C. et al. Clinical Practice Guidelines for Clostridium difficile Infection in Adults and Children: 2017 Update by the Infectious Diseases Society of America (IDSA) and Society for Healthcare Epidemiology of America (SHEA). Clin Infect Dis 66, e1-e48 (2018).
  2. Ford, D. C., Schroeder, M. C., Ince, D. & Ernst, E. J. Cost-effectiveness analysis of initial treatment strategies for mild-to-moderate Clostridium difficile infection in hospitalized patients. Am J Health Syst Pharm (2018).
  3. Nelson, R. L., Suda, K. J. & Evans, C. T. Antibiotic treatment for Clostridium difficile -associated diarrhoea in adults. Cochrane Database of Systematic Reviews (2017).
  4. Goldenberg, J. Z. et al. Probiotics for the prevention of Clostridium difficile-associated diarrhea in adults and children. Cochrane Database Syst Rev 12, CD006095 (2017).
  5. Crobach, M. J. T., Baktash, A., Duszenko, N. & Kuijper, E. J. Diagnostic Guidance for C. difficile Infections. Adv Exp Med Biol 1050, 27-44 (2018).






The ABC’s of ABG’s or How to read a blood gas without the Hassel(bach)

A blood gas interpretation is often a fear inducing “pimp” question. Probably because there is a so much packed into them and at some point, some basic math is needed. So, let’s try to unpack it a little so we have more method and less madness. I’m going to divide this up into 4 parts: The explanation (Part I), The calculation of the ABG (Part II), The Differential Diagnoses (Part III), The Practice (Part IV)


Blood gases are ruled by the often cited but never remembered: Henderson-Hasselbach equation:Screen Shot 2018-07-08 at 5.30.47 PMI think its easier to remember written in the ABCD form:

Screen Shot 2018-07-08 at 5.30.52 PM

There are only FIVE RULES to understanding ABG’s

  1. The primary disorder causes the pH
  2. The primary disorder is moves with the pH in the resulted direction
    1. Metabolic disorders are like a boy band…They always changes in ONE DIRECTION (i.e pH goes in the same direction as the primary disorder)
    2. Credit to Joel Topf of the curbisders podcast for that dad joke…
  3. Compensation occurs in the same direction as the primary disorder
  4. The body can’t make a large number of anions so an anion gap always means a primary metabolic acidosis is present
  5. A second primary disorder exists when the compensation doesn’t completely correct for the problem


Screen Shot 2018-07-08 at 5.39.35 PM

The 4 steps in reading an ABG:

Step I: Determine the pH and Primary disorder:

–      If pH, HCO3, and pCO2 are ONE DIRECTION then primary disorder is metabolic

e.g. All down:    <7.40, CO2 <40, HCO3 <24 = metabolic acidosis

e.g. All up:     >7.40, CO2 >40, HCO3 >24 = metabolic alkalosis

–    If pH, pCO2, HCO3 are in opposite directions then primary disorder is respiratory

Step II: Determine if there is a gap acidosis. If there is, then a  gap acidosis must be present

  • Gap = Na (corrected for glucose) – (Cl + HCO3)
  • PEARL: Remember to use the bicarb from the BMP/CMP not the ABG!
  • An AG of >30 is very likely to have an AGMA
  • An AG 20-29 then clinically 1/3 will not have metabolic acidosis
    • The Bicarb is CALCULATED in the ABG and MEASURED in the BMP
      • The pCO2/HCO3 ratio should always be checked
        • H+ = 24 x pCO2/HCO3-
    • Na is falsely low in hyperglycemia and must be corrected to get the correct Na. To correct do the following: For every 100 over 100 glucose add 1.6 to Na
    • Some people correct the  gap for albumin but you probably don’t need to. However, if you did it would be:
      • Corrected gap = AG + [2.5x(4-albumin)]

Step III: Determine Compensation (occurs in the same direction as the primary disorder)

  • Remember pH is inversely related to pCO2 and directly proportional to HCO3
  • If the HCO3 is low then the PCO2 should lower to compensate
  • See Compensation question below
    • Some say if the last two digits of the pH = pCO2 then NO respiratory disturbance occurs (eg. pH = 7.40 and pCO2= 40 then no respiratory disturbance)

Step IV:  Calculate the excess (or Delta) Gap (that is take out the GAP):

  • (Anion Gap – 12) + HCO3
  • If Excess > 25 then underlying Metabolic Alkalosis
  • If Excess < 23 then underlying Non-gapacidosis


Delta Gap =∆AG-∆HCO -=Na+-(Cl++HCO -)-12-(24-HCO -)

=Na+-Cl – 36

If the DG is significantly positive (>+6), a metabolic alkalosis  (IN ADDITION TO AGMA) is present because the rise in AG is more than the fall in HCO3-.

Conversely, if the DG is significantly negative (<-6), then a hyperchloremic (non-gap) acidosis (IN ADDITION TO AGMA) is present because the rise in AG is less than the fall in HCO3-.

*You could stop at the above step at get most of the way there*


Step IVa: Calculate “correction equations” to find the second primary disorder

  • Is there ENOUGH compensation to make up for the primary disorder (Is there a SECOND PRIMARY disorder?) ∆= Delta = Change
  • Correction equations can be made into a mnemonic (not a great one but kinda) if you remember things alphabetically (metabolic then respiratory, acidosis then alkalosis, acute then chronic) and the numbers  1.5 – 8 = 7, 1,2,3,4:
    • Metabolic acidosis:           pCO2 =  1.5 × HCO+ 8 ± 2 (Winter’s formula)
    • Metabolic alkalosis:           ∆ pCO2 = 9[∆ HCO3] OR                                         (pCO2 =0.9x HCO3+9±5) [Narins]
    • (Acute) Respiratory acidosis:   pCO2:HCO3 changes 10:1
    • (Acute) Respiratory alkalosis:  pCO2:HCO3 changes 10:2
  • (Chronic) Respiratory acidosis:         pCO2:HCO3 changes 10:4
  • (Chronic) Respiratory alkalosis:       pCO2:HCO3 changes 10:3
  • Alternately remembered as: The RESPIRATORY corrections table 
  pCO2 : HCO3
  Acidosis Alkalosis
Acute 10:1 10:2
Chronic 10:4 10:3


STEP I:   LOOK AT THE PH (>7.40 is alkalosis, <7.40 is acidosis)
if >6 there is a metabolic alkalosis)



         Old: CAT MUDPILES                                                       New: GOLDMARK

C CO, CN   G Glcyols (ethylene and propylene)
A AKA   O 5-oxoproline (Pyroglutamic Acid) [from chronic acetaminophen toxicity]
T Toluene   L L-Lactic acidosis
M Methanol   D D-Lactic acidosis (short gut syndromes)
U Uremia   M Methanol
P PARALDEHYDE, Pyroglutamic Acid, Phenphormin, Paraquat, Propylene Glycol   R Renal Failure
I INH, Fe, Ibuprofen (large doses)   K Ketosis (DKA/AKA)
L Lactate      
E Ethylene glycol      
S Salicylates      

H Hyperchloraemia
A Acetazolamide, Addison’s
D Diarrhea from ileostomies, fistulas


Use urinary anion gap [= (Na+ + K+) – Cl-] to differentiate between GI and renal causes

The remaining significant ions are NH4+ or  HCO3-

Renal causes increase HCO3- excretion thus increased urinary AG

GI causes increase NH4+ excretion thus decreased urinary AG



Screen Shot 2018-07-08 at 5.30.21 PM

3-L Lytes (Ca,K, Na, Mg), Lipids, Lithium
A Albumin
M Multiple Myeloma (IgG – cationic; IgA is anionic)
B Bromide, polymyxin B
  • Analytical errors like increased Na+ (most common), increased viscosity, iodide, increased triglycerides)
  • Decrease in anions (albumin, dilution)
  • Increase in cations (multimyeloma (IgG – is a cation; IgA is an anion), hyperkalemia, hypercalcemia, hypermagnesemia, lithium, polymixin B)
  • Bromide OD (causes falsely elevated chloride measurements)


Alkaline Input

  • Bicarbonate Infusion
  • Hemodialysis
  • Calcium Carbonate
  • Parenteral Nutrition

Proton Loss

  • GI Loss (vomiting, NG suction)
  • Renal loss
  • Diuretics
  • Mineralocorticoids


C CO2 overproduction (Malignanty Hyperthermia) or CNS Depression (Trauma or Toxins)
L Lung obstruction/injury (Upper or Lower)
I Inadequate ventilation
M Myopathies
B OBesity – Pickwickian syndrome

 6. RESPIRATORY ALKALOSIS: (Only 2 general causes)

Stimulated Respiratory Drive

– Hypoxemia



Step I: Determine the pH and Primary disorder:

  • If pH, HCO3, and pCO2 are ONE DIRECTION then primary disorder is metabolic

Step II: Determine if there is a gap acidosis.

  • Gap = Na (corrected for glucose) – (Cl + HCO3)

Step III:  Calculate the excess (or Delta) Gap:

  • Na – Cl – 36
  • If Excess > 6 then underlying Metabolic Alkalosis
  • If Excess < 6 then underlying Non-AGMA

Step IV: Determine Compensation (same direction as the primary disorder)


Practice Problems:

  1. pH 7.50 / pCO2 20 / HCO3 15 / Na 140 / Cl 103
  2. pH 7.40 / pCO2 40 / HCO3 24 / Na 145 / Cl 100
  3. pH 7.10 / pCO2 50 / HCO3 15 / Na 145 Cl 100
  4. pH 7.37 / pCO2 18 / HCO3 10
  5. pH 7.50 / pCO2 48 / HCO3 36
  6. pH 7.35 / pCO2 56 / HCO3 30
  7. pH 7.56 / pCO2 22 / HCO3 23
  8. pH 7.14 / pCO2 18 / HCO3 8 / Na 134 / Cl 104
  9. pH 7.45 / pCO2 17 / HCO3 12 / Na 139 / Cl 114

Practice Answers (Primary disorder is listed first)

  1. Respiratory Alkalosis and Anion Gap Metabolic Acidosis (e.g. aspirin overdose)
  2. Gap Acidosis AND metabolic alkalosis (e.g. A vomiting renal failure patient)
  3. Primary Respiratory alkalosis, Gap Acidosis AND metabolic alkalosis
  4. Metabolic acidosis, predicted pCO2 = 23, Respiratory alkalosis
  5. Metabolic alkalosis, pCO2 48
  6. Respiratory acidosis HCO3 acute: 26, HCO3 chronic 29, Metabolic alkalosis
  7. Respiratory alkalosis, HCO3 acute: 20, HCO3 chronic 16, Metabolic alkalosis
  8. Metabolic acidosis that is a gap acidosis with an additional non-gap acidosis
  9. Respiratory alkalosis with both a gap and non-gap metabolic acidosis.


1: Narins RG, Emmett M. Simple and mixed acid-base disorders: a practical
approach. Medicine (Baltimore). 1980 May;59(3):161-87. PubMed PMID: 6774200.

2:  Baillie JK. Simple, easily memorised ‘rules of thumb’ for the rapid assessment of physiological compensation for respiratory acid-base disorders. Thorax 2008;63:289-290 doi:10.1136/thx.2007.09122

3. Haber RJ. A practical approach to acid-base disorders. West J Med. 1991
Aug;155(2):146-51. Review. PubMed PMID: 1843849; PubMed Central PMCID:

A “Simplified” DKA algorithm and its rationalization

DKA Algorithm Lucid Chart



Diabetic Ketoacidosis is a life-threatening condition with the possibility of cerebral edema that occurs in type 1 Diabetes Mellitus (T1DM) and occasionally in type 2 DM (T2DM). For many it is a difficult process to manage with many moving targets, time frames, and life-threatening consequences if mistakes are made. No wonder these people go to the ICU. They need 1:1 care. But they all start out in the ED and we HAVE to make this diagnosis. That’s why on this deep dive I want to go into the literature behind the algorithms. This way you will see why I wrote up my own Simplified DKA algorithm “A 3-pronged approach” (see above). The mantra governing this “sweet” DKA treatment algorithm is “DKA took a bit to happen, so you should take a bit to treat it” (aka don’t rush DKA!)

Just as with any diagnosis that has been around since the dawn of time, the literature for DKA is…well nearly non-existent. I mean it’s bad. Yes, there are a lot of little studies, but seriously shouldn’t there be some big RCT’s? I mean look at cardiology literature those guys have thousands of patients enrolled! It’s so bad that Tran1 in her article entitled “Review of Evidence for Adult Diabetic Ketoacidosis Management Protocols” concluded “there is a major deficiency of strong evidence for the optimal management of DKA” and that “all components of DKA management would benefit from prospective RCTs “. Tell us how you really feel, why don’t ya! Despite this paucity of the literature we will go through what we can.


Oh, geez are you going to tell me there is controversy in DIAGNOSING DKA? No…no, no, no…well yes.  As a side note, I love America but I was born in Canada, so every once in a while, I think “God save the queen” because occasionally those Brits and Canucks have a good idea. This might just be the case in the diagnostic criteria of DKA.  The table below, adapted from US guidelines 2,  shows how “we the people” define DKA:

DKA Diagnostic criteria

How do they do it in the UK 3? Well it’s literally a mnemonic of the letters DKA. Oh, come on! Dammit, that’s good! Even I can remember that its:

D – “Dextrose”: 200 mg/dL or previous history of DM

K – “Ketones”: Serum ketones >3 mmol/L or >2+ on urine ketone stick

A – “Acidosis”: pH <7.3, Bicarb <15 mmol/L, Gap (no cutoffs listed)

Well “Ello Gov’nor”! I hate to admit it but that’s good. Come one ADA! Just adopt that one… too easy! You may also notice that this definition is a bit more conservative and uses a cutoff lower than that of the U.S (250 mg/dL). Obviously, this increases capture rate (sensitivity) of DKA and may also enhance identification of euglycemic DKA.  Euglycemic DKA is thought to occur in 10% of patients with DKA3. This can occur with certain medications such as the sodium-glucose cotransporter inhibitor, Invokana [canagliflozin] 4.  This can also occur in pregnancy. Pregnancy is a state of insulin resistance thus insulin requirement progressively rises throughout pregnancy explaining the higher incidence of DKA in the second and third trimesters 5. In addition to euglycemia other conditions that make DKA difficult to diagnose include:

  1. Conditions that increase serum bicarbonate (e.g. vomiting)
    1. Can cause a mixed acid/base disorder so pH is not as low
  2. Causes of significant osmotic diuresis (excessive protein intake)
    1. Loss of ketoacids may lead to a normal gap
  3. AKA which can cause elevated β-hydroxybutyrate level
    1. See section on differential diagnosis.

Take home point:  Screen all patients with point of care glucose testing >200 mg/dL for DKA. Know the definition of DKA. Consider euglycemia and other co-morbidities that make DKA more difficult to diagnose.


Obviously, there are a lot of different ways to do the same thing.  It would be great to have a non-invasive method of screening for DKA. End-tidal CO2 has been postulated as one such way to rule out DKA. Sadly, there just isn’t enough literature yet. One such trial of 21 DKA patients, showed that an ETCO2 > 35 mmHg was 100% sensitive. In the table below I’ve listed the studies (an “half-asked” look at the literature on the topic, so I may have missed some). If you look at that table there really aren’t any consistencies at all in the value of ETCO2 used, the correlation with bicarb or pH and so on. I’d like to see a large study that finds a value of ETCO2 that is 100% sensitive. Possibly, with shared decision making, using a value greater than 36 mmHg might be reasonable, assuming easy access to ETCO2. This may be an option but without larger trials I’m not sure this idea is ready hang my hat on!

Screen Shot 2018-07-01 at 4.35.17 PMGETTING GAS-SY: ARTERIAL BLOOD GAS (ABG) VS. VENOUS (VBG)

So, if a pH is required to make the diagnosis, then we probably need a blood gas. Can we do a VBG instead of an ABG? Afterall, it just can’t feel good poking a needle into a radial artery. The US guidelines6 allow for venous pH in those “without heart, lung, or kidney disease”.  How good is the correlation between venous and arterial pH? It seems pretty darn good. There are multiple studies that all show similar findings. This 2014 systematic review of 1747 acidotic patients from 15 studies showed a that the venous pH was only 0.03 (95%: ±0.004), lower than the arterial pH7. This observational prospective trial of 200 DKA patients showed similar results with a venous pH being 0.015 (95%: ±0.006) lower than arterial pH8. The same study also showed that the arterial pH (when drawn simultaneously to venous) only changed the treatment or disposition in 5/200 patients or 2.5% of the time8.

Unfortunately, the remaining components aren’t that great but do you really need those components to manage DKA? Probably not. I will say that this systematic review by Bloom found in 392 patients a venous pCO2 of < 45 mmHg had a 100% sensitivity and negative predictive value for an arterial pCO2 to be not greater than 50 mmHg 9. That is, if the vbg shows normocapnia then so will the abg.

Take home point: The venous pH will be 0.01 lower than the arterial pH and a pCO2 of < 45 mm Hg will have a normal arterial pCO2


            This is probably is probably the most contentious area of DKA. Its the age old balanced fluids (e.g LR) vs normal saline (NS) debate. I’m guessing the answer is probably that the fluid type doesn’t matter that much. For now, I think the discussion of what fluid to use in the critically ill patient will be reserved for another blog. However, there are a few caveats and quid pro quo’s that I’d like to discuss briefly (“You keep using that word. I do not think it means what you think it means …”).

            As Tran points out in her review the ideal solution to use is “unclear” and most of the studies in DKA are pretty small1. Probably the biggest concern with using Normal Saline is a hyperchloremic metabolic acidosis. Additionally, there is a likely sub-clinical change in renal function that has yet to result in any meaningful clinical outcome such as ICU days or dialysis days. Recently, however, there has been a large trial published specifically looking at normal saline in DKA. This landmark NEJM trial by Kupperman compared 0.9%NS to 0.45% NS in 1389 events of DKA 10. While he found no clinical differences in the primary end point (neurologic outcomes) he did find some statistically significant electrolyte differences in the form of hypophosphatemia, hyperchloremia and hypocalcemia. However, none of these abnormalities changed the speed of recovery from DKA as measured by time to resolution of the DKA (~14 hrs) or time to discharge (~46 hrs). For this reason, in my “3-pronged approach” to DKA, I elected to go with 0.45% NS for everything except boluses since there is no difference in recovery time and less laboratory abnormalities.

Take home point: Use NS to bolus then switch to 0.45%NS after. There does not appear to be any benefit at this time of “balanced fluids” vs 0.45%NS


            Typically, the total body water (TBW) deficit in DKA is thought to be around a 10% loss and is, therefore, the volume replaced. However, I think this is also up for debate. In 1989 Adrogue11 looked at “large volumes” (1000 ml/hr over 4 hours) vs “small volumes” (500 ml/hr over 4 hours) of fluid replaced. They found NO difference in resolution of DKA but did show less cost, less fluid infused and less hyperchloremia with the smaller volumes.  Similarly, Kupperman10 also looked at large (10% TBW loss) vs small (5%TBW loss) replaced fluid volumes. They too found less hyperchloremia in smaller volumes and no differences in outcomes or resolution of DKA. For this reason, in my “3-pronged approach” to DKA I elected to go with bolusing 10 ml/kg as well as calculating and replacing only a 5% TBW deficit since there is no difference in recovery time, while allowing for a benefit of less cost and less laboratory abnormalities. Obviously, the patient in shock will require appropriate fluid boluses to whatever endpoint of resuscitation the clinician uses. However, this is not the case for most DKA.

Take home point: Replace less fluid unless there is SHOCK. Use a TBW deficit of 5% and use 10 ml/kg boluses initially unless there is SHOCK.


            As is the case in most of medicine we have persistent dogma that has more to do with “one time I saw this happen so that must be the case” versus a true cause-effect relationship. While the concern for cerebral edema is real in DKA, it is unlikely a result of the rate of fluid administration. In fact, this isn’t a new finding. In 2002 Azzopardi12 looked at 10 adults with DKA and treated them with isotonic or hypotonic solutions and then performed a CT scan before treatment and then at 12 hrs, 24 hrs and 6 months after initiation of treatment. They did not find clinical cerebral edema in either group. In 2013 Glaser13 showed that in 10 children randomized to get slow or fast infusions of fluid. They performed diffusion weighted MRI during and after treatment to see if there was vasogenic cerebral edema. They showed no MRI differences in children with DKA, regardless of rate of infusion. In 2018, Kupperman10 showed no difference in neurologic outcomes in 1389 events of DKA whether they were randomized to one of four groups with varying rates of infusion and normal vs half-normal saline.

            So what are some risk factors for the progression in children to cerebral edema? Glaser and Kupperman 14 found in 2001 that children with DKA vs random controls are at higher risk for cerebral edema if they have a low paCO2, a high BUN or are treated with bicarbonate.

Take home point: Cerebral edema is likely associated with bicarbonate and lower paCO2. The rate and fluid are probably not a cause of cerebral edema but I’m sure using less fluid overall is a good idea.


            You will notice in my Simplified DKA algorithm “A 3-pronged approach” there are two commonly included elements missing. The first is the purposeful omission of bicarbonate supplementation (the second is omission of phosphate replacement). We just said from above that bicarb treatment is a risk factor for cerebral edema. Subsequently, there are those that will make the argument that in very low pH <6.9 bicarbonate may have a role. The ADA guidelines2 suggest “Because severe acidosis may lead to [sic] a numerous adverse vascular effects, it is recommended that adult patients with a pH 6.9…receive…sodium bicarbonate…until the venous pH is 7.0…”. I disagree with this, but don’t just take my word for it. The UK guidelines3 also recommend against the use of bicarbonate replacement “with the rationale that fluid and insulin replacement alone will be sufficient to raise pH”. This is also demonstrated in several studies. This study looked at 147 pediatric DKA patients including 9 with a pH less than 7 and one with a pH of 6.73! Not only did they find no benefit, they found longer hospitalizations in the group given bicarb. In this 1986 RCT of 20 adults with a pH from 6.9 to 7.14, bicarbonate therapy did not affect the speed of recovery in DKA. There seems to be a theme here and that’s enough for me. I might have to wear a redcoat from now on when I’m treating DKA but it seems worth it!

Take home point: There is no role for bicarbonate therapy in DKA even with a pH <7.0.


            As mentioned above phosphate is the other element that is absent from my Simplified DKA algorithm “A 3-pronged approach”. The ADA guideline recommends, in my opinion, a complex phosphate replacement. However, they only recommend it in patients with cardiac dysfunction, anemia, respiratory depression, or patients with phosphate levels <3.2mmol/L 2. The review by Tran states that “Prospective randomised studies demonstrated that phosphate replacement offers no improvement to DKA outcomes1. The NICE guidelines recommend to “not generally use phosphate replacement in the management of DKA in adults.15”.

Take home point: It’s highly unlikely you will need to replace phosphate. If you do, it should only be when <3.2 mmol/L or if respiratory depression. There is plenty of time to look this up and do it later.


            Notice how much we’ve discussed and HAVEN’T even spoken of insulin yet? Despite total-body potassium depletion, mild-to-moderate hyperkalemia is common in patients with hyperglycemic crises2. Since understanding this is critical I want to break this down into simple terms. This will be anything but a biochemistry lecture. Once you give insulin therapy you help get rid of the ketones and thus the acid in the blood. Less acid means you are further driving potassium into the cells. [Remember in hyperkalemia we give bicarb, a base (aka less acid) to drive K into the cells, same principal] If the K is already super low (<3.3) you can’t give insulin yet so you have to hold ‘the base” (insulin) so you don’t push more K into the cell and cause a dangerously low potassium. Therefore, small amounts of K are added until the K is >3.3. If the k is 3.3 to 5.3 then all you need to do is to maintain that k balance so add 20-30 mEq to each liter of maintenance fluid. If K is >5.3 then it will come down on its own just by treating the acidosis. PEARL: Make sure urinary flow is appropriate in DKA otherwise you can’t urinate out the K that is moving around.

Take home point: Insulin therapy will LOWER the serum potassium. Don’t give insulin if its already low because you will cause a dangerously low K. Try to keep the K between 3.3 and 5.3. Monitor for good urine output


So lastly, (like I said “no rush”) is insulin. I don’t think there is much to say about this. Regular insulin is used because of its low cost. A bolus does not change time to resolution of DKA, length of hospital stay, hypokalemia, or other complications including death 1.  In 2008, a small prospective randomized trial found that an initial bolus of insulin avoided the need for supplemental insulin doses if the insulin infusion rate was at least 0.14 units/ kg/h 16.


            With any case of a metabolic acidosis one should keep a broad differential of its causes. One differential to consider is alcoholic ketoacidosis (AKA). Sometimes the two can be difficult to distinguish. This study compared 12 patients with DKA and 12 with AKA. They found on average the following serum value similarities in DKA vs. AKA (respectively):  bicarb [11 vs 11 mEql/L], venous pH [7.18 vs 7.28], anion gap (30 vs 30), Sodium (134 vs 139 mEq/L), beta-hydroxybutyrate (7 vs 6 mmol/L). They found on average the following serum value differences in DKA vs. AKA (respectively): glucose (578 vs 118), acetoacetate (2.6 vs 1.1), beta-hydroxybutyrate to acetoacetate ratio (3:1 vs 7:1), and lactate (1.6 vs 3 mmol/L) 17. Thus, the biggest differentiator was the ketone ratio of 3:1 in DKA and 7:1 in AKA and not the presence of beta-hydroxybutyrate.


            That’s sums up the understanding and evidence of DKA. Any protocol you use is better than none1 so make sure you have a protocol easily available and read it over BEFORE you have a case of DKA. Until next time…

DKA Algorithm Lucid Chart






  1. Tran, T. T. T. et al. Review of Evidence for Adult Diabetic Ketoacidosis Management Protocols. Front Endocrinol (Lausanne) 8, 106 (2017).
  2. Kitabchi, A. E., Umpierrez, G. E., Miles, J. M. & Fisher, J. N. Hyperglycemic crises in adult patients with diabetes. Diabetes Care 32, 1335-1343 (2009).
  3. Dhatariya, K. K. & Vellanki, P. Treatment of Diabetic Ketoacidosis (DKA)/Hyperglycemic Hyperosmolar State (HHS): Novel Advances in the Management of Hyperglycemic Crises (UK Versus USA). Curr Diab Rep 17, 33 (2017).
  4. Kelmenson, D. A. et al. Euglycemic diabetic ketoacidosis with prolonged glucosuria associated with the sodium-glucose cotransporter-2 canagliflozin. Journal of investigative medicine high impact case reports 5, 2324709617712736 (2017).
  5. Kamalakannan, D. Diabetic ketoacidosis in pregnancy. Postgraduate Medical Journal 79, 454-457 (2003).
  6. Kitabchi, A. E. et al. Management of hyperglycemic crises in patients with diabetes. Diabetes Care 24, 131-153 (2001).
  7. Byrne, A. L. et al. Peripheral venous and arterial blood gas analysis in adults: are they comparable? A systematic review and meta-analysis. Respirology 19, 168-175 (2014).
  8. Ma, O. J., Rush, M. D., Godfrey, M. M. & Gaddis, G. Arterial blood gas results rarely influence emergency physician management of patients with suspected diabetic ketoacidosis. Academic Emergency Medicine 10, 836-841 (2003).
  9. Bloom, B. M., Grundlingh, J., Bestwick, J. P. & Harris, T. The role of venous blood gas in the emergency department: a systematic review and meta-analysis. Eur J Emerg Med 21, 81-88 (2014).
  10. Kuppermann, N. et al. Clinical Trial of Fluid Infusion Rates for Pediatric Diabetic Ketoacidosis. N Engl J Med 378, 2275-2287 (2018).
  11. Adrogué, H. J., Barrero, J. & Eknoyan, G. Salutary effects of modest fluid replacement in the treatment of adults with diabetic ketoacidosis: use in patients without extreme volume deficit. Jama 262, 2108-2113 (1989).
  12. Azzopardi, J., Gatt, A., Zammit, A. & Alberti, G. Lack of evidence of cerebral oedema in adults treated for diabetic ketoacidosis with fluids of different tonicity. Diabetes Res Clin Pract 57, 87-92 (2002).
  13. Glaser, N. S. et al. Subclinical cerebral edema in children with diabetic ketoacidosis randomized to 2 different rehydration protocols. Pediatrics 131, e73-80 (2013).
  14. Glaser, N. et al. Risk factors for cerebral edema in children with diabetic ketoacidosis. The Pediatric Emergency Medicine Collaborative Research Committee of the American Academy of Pediatrics. N Engl J Med 344, 264-269 (2001).
  15. Amiel, S. A., Pursey, N., Higgins, B., Dawoud, D. & Guideline, D. G. Diagnosis and management of type 1 diabetes in adults: summary of updated NICE guidance. BMJ 351, h4188 (2015).
  16. Kitabchi, A. E., Murphy, M. B., Spencer, J., Matteri, R. & Karas, J. Is a priming dose of insulin necessary in a low-dose insulin protocol for the treatment of diabetic ketoacidosis. Diabetes Care 31, 2081-2085 (2008).
  17. Umpierrez, G. E. et al. Differences in metabolic and hormonal milieu in diabetic- and alcohol-induced ketoacidosis. J Crit Care 15, 52-59 (2000).